Participant Presentations and Session Transcript for May 8

1

1	UNITED STATES FEDERAL TRADE COMMISSION
2	and
3	UNITED STATES DEPARTMENT OF JUSTICE
4
5
6
7	SHERMAN ACT SECTION 2 JOINT HEARING
8	UNDERSTANDING SINGLE-FIRM BEHAVIOR:
9	CONDUCT AS RELATED TO COMPETITION
10	TUESDAY, MAY 8, 2007
11
12
13
14
15	HELD AT:
16	UNITED STATES FEDERAL TRADE COMMISSION
17	HEADQUARTERS BUILDING, ROOM 432
18	600 PENNSYLVANIA AVENUE, N.W.
19	WASHINGTON, D.C.
20	9:00 A.M. TO 1:00 P.M.
21
22
23
24	Reported and transcribed by:
25	Brenda Smonskey

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

2

1	MODERATORS:
2	DEBORAH PLATT MAJORAS
3	Chairman
4	Federal Trade Commission
5	and
6	TOM BARNETT
7	Assistant Attorney General
8	Antitrust Division, U.S. Department of Justice
9
10	PANELISTS:
11
12	Susan Creighton
13	Jeff Eisenach
14	Tim Muris
15	Bob Pitofsky
16	Doug Melamed
17	Jim Rill
18	Charles F. (Rick) Rule
19	Greg Sidak
20
21
22
23
24
25

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

3

1	C O N T E N T S
2
3	Introduction........................................................................................................................................................... 4
4
5	Moderated Discussion......................................................................................................................................... 13
6
7	Conclusion.........................................................................................................................................................162
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

4

1	P R O C E E D I N G S
2	- - - - -
3	CHAIRMAN MAJORAS: Good morning, everyone.
4	Welcome to this final wrap-up panel of the
5	hearings that we, the FTC, together with the DOJ
6	Antitrust Division have been holding over the course
7	of almost the past year.
8	I'm delighted to be here today to moderate
9	this final session with my very good friend and
10	colleague, Tom Barnett, Assistant Attorney General
11	for the Antitrust Division.
12	So I thank you all for being here. I also
13	thank our panelists for taking the time away to be
14	with us this morning.
15	Before I get started, I should ask all of
16	you just as a courtesy that if you have anything on
17	that rings or otherwise makes noise, if you could
18	turn off at least that part of it. We would
19	appreciate it.
20	We ask that you not make comments, at least
21	not above your breath, during the session or yell
22	out questions from the audience, please.
23	I want to start this morning by thanking the
24	staff from the FTC and from the Department of
25	Justice Antitrust Division for their incredible work

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

5

1	over the course of the last year in putting together
2	27 Section 2 hearing sessions over the course of the
3	year.
4	These things have gotten to the point where
5	I think they go so well and so smoothly that you
6	forget how much work is going on behind the scenes.
7	But I see Pat here and Bill Cohen and Gail.
8	They can tell you all the work that has gone on
9	behind the scenes. We are truly grateful for their
10	contributions.
11	I also want to express my appreciation to
12	the 130 panelists we have had over the course of
13	these sessions. They have made an incredible
14	contribution to these hearings.
15	I wanted to convene the hearings because it
16	seemed to me that the debate over where we should be
17	drawing the permissible lines for conduct by firms
18	with market power needed something of a boost.
19	I was a little bit worried that it might be
20	getting stuck. It seemed like we were drawing
21	lines, to be sure, but we were drawing more like
22	battle lines around certain tests or certain
23	arguments.
24	And our hope was that through these hearings
25	we could identify or highlight areas certainly of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

6

1	broad consensus in enforcement against single-firm
2	conduct and then also draw out the areas that
3	require further rigorous analysis and guidance.
4	So starting with the opening session on June
5	20th, we have held hearings on a wide range of
6	conduct, from predatory pricing to exclusive dealing
7	to bundled and loyalty rebates and the whole
8	spectrum, as well as sessions on monopoly power,
9	remedies, market definition.
10	We also held a session on empirical
11	research, during which we heard about the research
12	that exists on Section 2 areas as well as areas
13	where further research would be helpful.
14	We held a session on international
15	perspectives, where we heard from a number of
16	foreign competition agency officials as well as
17	practitioners and academics in the field.
18	We held a session on business history in
19	which we examined some of the more important
20	monopolization cases of the past century.
21	We had a session on business strategy so we
22	could learn more about what business schools are
23	teaching future business leaders and executives,
24	what they are teaching them and how that could
25	ultimately impact competition and conduct.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

7

1	I had hoped, as you all know, from the very
2	beginning that we could get a fair amount of input
3	from the business community so we could actually
4	really think about certain types of conduct, why
5	folks are engaged in it.
6	And I was pleased that we were able to hold
7	two out of town hearings this time, get outside the
8	Beltway. We held a hearing in Berkeley, California
9	and Chicago, Illinois, which I was very pleased
10	about.
11	Through all this, we have endeavored to
12	select panelists that could provide a wide diversity
13	for us of viewpoints on these important topics.
14	So here we are. We are at the last
15	roundtable discussion. We held another almost last
16	roundtable discussion last week. So here we are
17	today.
18	We will ask our panelists to comment on a
19	wide range of issues.
20	We will not have speaker presentations
21	today. We will get directly into questions from our
22	panelists, which we thought would be a richer forum
23	to take advantage of the great wisdom and experience
24	of this distinguished panel.
25	With that, I will tell you -- I think you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

8

1	probably know who they are, but I'm going to tell
2	you.
3	I will start with four of the panelists who
4	I will introduce. Tom will introduce the others.
5	I will introduce all the former FTC folks,
6	and Tom will introduce the former DOJ folks plus
7	one.
8	I was thinking what we might do is have them
9	duke it out. Maybe we can solve all the problems.
10	We have a new form of clearance agreement of some
11	sort.
12	So to my far right is Susan Creighton.
13	Susan is a partner at the Wilson Sonsini firm after
14	having served here as the director of the FTC's
15	Bureau of Competition, and it has been my great
16	pleasure to work with Susan.
17	Susan is quite well known in this area of
18	Section 2 law and in particular of late in the area
19	of cheap exclusion.
20	So we will look forward to her comments
21	today.
22	Jeff Eisenach is the chairman of Criterion
23	Economics and adjunct professor at the George Mason
24	School of Law.
25	He has served in senior policy positions at

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

9

1	the FTC and also at the Office of Management and
2	Budget. He was a cofounder of the Progress and
3	Freedom Foundation. And he is also someone willing
4	to play golf with me.
5	Tim Muris -- I can't introduce Doug because
6	he used to be at DOJ. Sorry, Doug. So did I.
7	Tim Muris will be here. We knew that he
8	would have to be a little bit late today. I will go
9	ahead and introduce him anyway.
10	He is a George Mason University Foundation
11	professor of law, of counsel at O'Melveny & Myers
12	and a co-chair of that firm's antitrust practice.
13	He also, of course, served as chairman of
14	the FTC until 2004. And in his previous life in the
15	'80s was director both of the Bureau of Competition
16	and the Bureau of Consumer Protection.
17	Tim will be here later this morning.
18	Finally, to Tom's left we have Bob Pitofsky,
19	the Joseph and Madeline Sheehy professor in
20	antitrust and trade regulation law at Georgetown
21	University Law Center, where he formerly served as
22	dean.
23	He is also counsel at Arnold & Porter and
24	formerly chairman of the FTC, prior to Tim Muris, of
25	course.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

10

1	We have a lot for which we are grateful to
2	Bob, but one I think is that Bob really
3	reinvigorated this concept of hearings at the FTC
4	during his tenure.
5	That, of course, is the tremendous legacy
6	that brings us here today. So thank you.
7	Now I would like to turn it over to Tom
8	Barnett.
9	MR. BARNETT: Thank you, Debbie.
10	I also would like to underscore my thanks to
11	the staff, who have worked very hard. And in some
12	sense it seems like yesterday, it was almost a year
13	ago when Debbie and I stood up, I think over in that
14	corner of the room, along with a few other people
15	and helped launch these hearings.
16	But to the staff I have a feeling that may
17	seem like about 10 years ago, given the number of
18	sessions and panelists and issues.
19	As we were working through the preparation
20	for the hearing today, one of the things that really
21	struck me is the range of issues and the depth of
22	thought that has gone into preparing each and every
23	one of these sessions.
24	I know it is a tremendous amount of time and
25	effort. But I also agree with Debbie that this is

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

11

1	an extraordinarily important topic.
2	I have long viewed this, along with I think
3	Judge Posner who said this as well, really to be the
4	most challenging area of antitrust enforcement in
5	many ways, because large dominant firms can impose
6	very significant costs in terms of consumer welfare.
7	It is also the most difficult area in which
8	to avoid making mistakes as a government enforcer,
9	both in terms of condemning conduct that actually
10	can be beneficial, and even if you find a problem,
11	in crafting remedies that will fix the problem
12	without doing more harm than good.
13	And while I do agree that there are many
14	areas of consensus at least within the United States
15	in this area -- and I think the hearings have done a
16	good job of highlighting some of those things -- I
17	also think there are some very important issues that
18	remain open.
19	And I'm optimistic with the wide range of
20	experience and talent that we have had, the benefit
21	of economists, lawyers, business people, academics,
22	and certainly with the degree of experience and
23	wisdom we have at the panel here today, I expect we
24	will have resolved all of this by 1:00 today.
25	With that, I do want to move toward the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

12

1	discussion, which we have a lot to cover in a lot of
2	very -- it seems like a long time, but I have a
3	feeling it will go quickly.
4	So let me just move to the introductions.
5	I will start off with introducing Doug
6	Melamed, who is a partner and co-chair of
7	WilmerHale's -- do you say WilmerHale?
8	MR. MELAMED: I am supposed to.
9	MR. BARNETT: -- antitrust and competition
10	department and former Deputy Assistant Attorney of
11	the Department of Justice's Antitrust Division,
12	where he had a little bit of experience in some
13	Section 2 matters.
14	And then over to my left is Jim Rill, who
15	I'm sure everyone knows, who is a partner at Howrey
16	and the former Assistant Attorney General of the
17	Antitrust Division.
18	To his left is Rick Rule, who is a partner
19	at Cadwalader, Wickersham & Taft and also a former
20	Assistant Attorney General at the Antitrust
21	Division.
22	And down at the left is Greg Sidak, who is a
23	visiting professor of law at Georgetown University
24	Law Center and a founder of Criterion Economics.
25	He served as the deputy general counsel of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

13

1	the FCC and senior counsel and economist to the
2	Council of Economic Advisors over in the executive
3	branch.
4	So welcome to everyone. And with that I say
5	why don't we get to it.
6	In terms of format, Debbie and I thought we
7	would basically play tag team in terms of who will
8	lead off each topic, with the idea, however, that
9	each of us will jump in as seems useful.
10	And we are going to start off with the first
11	topic being general standards and issues.
12	I will ask the very first question in the
13	broadest possible form, which is I would like to ask
14	which one or two issues -- and I would ask no more
15	than two to keep it short -- that you think are the
16	biggest problems or concerns facing antitrust
17	enforcement today in the area of Section 2 that we
18	should try to address in the report that comes out
19	of this.
20	To start off, why don't I ask Jim Rill to
21	jump in.
22	MR. RILL: Thank you, Tom.
23	Let me say it is an extraordinary honor to
24	be here on this panel of august personages and to be
25	invited to participate.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

14

1	I think one issue stands out in a claim
2	being addressed in the report, and I emphasize
3	report, not necessarily guidelines, but an
4	analytical report -- hopefully with some sense of
5	conclusion and advocacy -- and that is the area of
6	bundled pricing and loyalty discounts.
7	The area has abounded in some confusion ever
8	since the LePage's-3M decision. There are several
9	court decisions on the way up that may add clarity
10	or possibly further confusion to the issue.
11	But trying to provide advice in that
12	particular area is daunting. I think that there are
13	a number of solutions out there, or at least
14	potential solutions out there as we get into more
15	the merits of the discussion today.
16	But I think those particular areas are ones
17	that really stand out above the others in looking
18	for a detailed analysis and what I would propose to
19	be a report, which I earnestly hope is forthcoming
20	as a results of these hearings.
21	MR. BARNETT: Thank you.
22	Bob, would you like to give us your
23	perspective?
24	MR. PITOFSKY: Thank you.
25	It is very similar to Jim.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

15

1	We talked about whether we could reach
2	consensus. I suspect the best chance we have of
3	reaching consensus is on the issue of what is the
4	most pressing set of issues facing antitrust, and I
5	think it is defining exclusionary behavior under
6	Section 2.
7	I think it is a set of issues that is most
8	confusing, hard to predict, hard to counsel, hard
9	for judges to deal with.
10	Some people will hold out for the Robinson
11	Patman Act, but I don't quite think that is really
12	the toughest set of questions.
13	And as we will discuss today, what sort of
14	rule should we build on? Is it the balancing test
15	that was unanimously adopted by the Court of Appeals
16	in Microsoft and echoed I think in Aspen, or these
17	unitary tests. We all know the balancing test has
18	its flaws in terms of unpredictability and
19	difficulty in implementing in the context of a legal
20	proceeding.
21	But should we look for a unitary test, which
22	people understandably and with my admiration have
23	tried to come up with -- sacrifice of profits,
24	driving out a less efficient competitor and so
25	forth.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

16

1	I will give away my bottom line right now.
2	I think the unitary tests, much as I admire the
3	creativity of them, don't work, do more harm than
4	good. And therefore, I would stick with the
5	balancing test.
6	But I think that's what a lot of our
7	discussion this morning should be directed toward.
8	MR. BARNETT: Doug?
9	MR. MELAMED: I think the most important
10	thing that can come out of these hearings would be
11	an explicit clarification or articulation of the
12	purpose of rules about exclusionary conduct.
13	I had occasion before coming today to look
14	through some of the summaries of the hearings that
15	you have held thus far. I haven't read all the
16	testimony. But I did look at the summaries.
17	I had the impression that it was like an
18	unbounded exercise for a public policy class at the
19	Kennedy School.
20	There are all sorts of people with all sorts
21	of views about how to address tying, exclusive
22	dealing, predatory pricing, whatever the topic is,
23	unstated often in the dialogue, and I think often
24	explaining the disagreements among the parties, were
25	differences in assumptions about the purpose of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

17

1	antitrust.
2	Is it consumer welfare? Is it total
3	welfare? Is it dynamic analysis? Is it static
4	analysis? And so forth.
5	This problem doesn't arise in cases of
6	collusion, because in these cases, I think both the
7	normative and the analytical converge on the
8	understanding that the issue is, does the
9	arrangement increase or decrease the output of the
10	parties to the agreement.
11	In exclusion cases, we are often dealing
12	with a trade-off between the efficiency benefits to
13	the defendant and the exclusionary impact on rivals.
14	And I think we don't have a clear understanding of
15	what the antitrust objective is dealing with that
16	trade-off.
17	My own view is that none of the sort of
18	economic factors mentioned above is a sufficient
19	statement of the objectives. If you look at the
20	cases, and I think the cases are wise in this
21	regard, you see, of course, Trinko, saying that
22	monopoly profits can be a good thing.
23	More important, I think, you see some of the
24	earlier cases, Grinnell and ALCOA, cases that say in
25	effect quite explicitly that, if a monopolist gains

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

18

1	his monopoly power by skill, foresight and industry,
2	that's okay.
3	Those cases embrace a normative proposition
4	that is very important to the fact that antitrust
5	has been supported by the political system in this
6	country for 120 years. That normative proposition
7	is that if the conduct is permissible, in some sense
8	defined without regard to its consequences, it's
9	okay.
10	So what we have to do on the conduct
11	element, exclusionary conduct, is to focus on the
12	quality of the conduct defined without regard to its
13	impact on consumer welfare or dynamic welfare or
14	whatever.
15	It happens, I believe, that if you do that,
16	you are adopting, at least if you do it the way I
17	would do it, what works out to be a very good proxy
18	in the real world, given the problems of
19	administrability and so forth, for achieving the
20	economic objectives.
21	In any event, I think you cannot focus just
22	on the economic objectives. You have to identify
23	clearly the normative objectives of exclusionary
24	conduct law.
25	CHAIRMAN MAJORAS: Anybody want to take that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

19

1	on in terms of whether that is enough, whether
2	looking at the conduct of the defendant rather than
3	the impact on consumers or competitors is adequate?
4	MR. PITOFSKY: I already said I'm
5	uncomfortable with that. It puts the focus in the
6	wrong place.
7	My concern is not the behavior of the
8	monopolist, the defendant. I thought antitrust laws
9	were designed to advance and I think the bottom line
10	is, consumer welfare.
11	If you are looking for consumer welfare, I
12	think it is relevant but not dispositive to know
13	what the intent of the monopolist is and what the
14	nature of its conduct is.
15	But I want to pick up that just because the
16	monopolist behavior is efficient or involves a
17		sacrifice of profit doesn't answer the question. I
18	want to know how anticompetitive it is with respect
19	to consumers.
20	I thought at least in this country consumer
21	welfare and not total welfare -- maybe you can
22	challenge it in academia, but as far as the courts
23	are concerned -- consumer welfare is what it's
24	about.
25	MR. MELAMED: Can I make a brief comment in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

20

1	response to Bob, just a question?
2	Bob, if a firm builds a better mousetrap and
3	as a result obtains enduring market power, and the
4	effect of the enduring market power is overall to
5	make consumers worse off than they would have been
6	if they never built the mousetrap, do you condemn
7	that conduct because --
8	MR. PITOFSKY: How do consumers come out
9	worse off in the face of a better mousetrap?
10	MR. MELAMED: My mousetrap is 5 percent
11	better than the incumbents', I drive the incumbents
12	all out of business; after they leave, I raise
13	prices 5 percent. It is easy to think of
14	hypotheticals where consumers are worse off.
15	MR. PITOFSKY: That's superior skill as far
16	as I'm concerned and I don't have any problem with
17	it. But it's not the typical case.
18	MR. BARNETT: I'm not sure we have so much
19	disagreement.
20	Rick, you want to jump in?
21	MR. RULE: Sure. I am for once to the left
22	of both Doug and Bob. And perhaps I wouldn't say it
23	is one of the few times, because I actually agree
24	with them a lot.
25	But I think I agree with Bob probably

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

21

1	wholeheartedly, I guess. I said this before.
2	I worry about the unitary approaches to
3	single-firm conduct. I think it creates a lot of
4	issues.
5	My own personal view is, as I said before, I
6	don't think the world would be a terrible place
7	without Section 2 of the Sherman Act, because I
8	think most of the conduct that is worthy of
9	condemnation can be attacked through various other
10	legal means.
11	So to me, I would say the biggest issue is
12	cabining Section 2 and focusing it.
13	The problem with the unitary standards is, I
14	think, they presume a sort of capability of
15	regulators and enforcers and courts to distinguish
16	efficient from inefficient conduct that just doesn't
17	exist.
18	I think that I have always been very
19	impressed by some of the writings of Judge
20	Easterbrook and particularly the limits of
21	antitrust.
22	And the fact is, if you look, I think,
23	historically at tests that put a burden on a
24	defendant to justify its conduct as efficient,
25	inevitably the courts find it very difficult to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

22

1	agree or to see an efficiency.
2	So I think the focus really ought to be on
3	whether or not there is exclusion, foreclosure, or
4	whatever you want to say of competition.
5	I don't think that is a sufficient condition
6	to condemn something, but I think it is necessary.
7	It may be that the foreclosure, or the
8	exclusion is due to the fact that there is a better,
9	more desirable mousetrap, and that is an efficiency
10	defense, and I think there ought to be allowed an
11	efficiency defense.
12	But I think that an absolutely necessary
13	condition is market power on the part of the
14	individual and exclusion of competition.
15	The last point that I would make that I
16	think is often left unsaid in these sorts of
17	discussions but I think is very important, when you
18	are talking about going after unilateral conduct and
19	you don't have an agreement, you don't have all the
20	issues that I think, quite rightly, warrant
21	antitrust enforcement when you are talking about an
22	agreement. When you are talking about going after
23	unilateral conduct, you are essentially talking
24	about the government regulating behavior of
25	individuals, maybe companies. But it is unilateral

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

23

1	action.
2	And there, I think, we as a society, given
3	the way we are organized, should be very concerned
4	not only about the adverse economic effects, the
5	false positives, but also about the impact on
6	liberty, on creativity, and on all of the benefits,
7	not only to the economy, but also to our political
8	life that individual freedom and liberty bring.
9	CHAIRMAN MAJORAS: Susan, you were going to
10	make a comment before Rick.
11	MS. CREIGHTON: That's all right. I can
12	encompass it in my remarks, which was I have sort of
13	a 1 and 2A and B. Hopefully that is not breaking
14	the rules.
15	So the first point and I think actually
16	maybe directly in contrast to Doug, the first thing
17	I would love to see come out of the report is an
18	affirmation that the principle that I think
19	underlies the rule of reason both for Section 1 and
20	Section 2, which is consumer welfare as sort of the
21	touchstone for our analysis, should be really the
22	governing principle in terms of what we adopt for
23	specific rules for conduct under Section 2.
24	I think, like Bob, I'm not saying we can
25	come up with a single unifying test that would cover

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

24

1	all that type of conduct. But I believe that we
2	should be assessing the particular tests that we
3	adopt with respect to particular conduct in terms of
4	whether or not it does maximize consumer welfare and
5	is consistent with the rule of reason.
6	So I would use something like the Microsoft
7	test as sort of our default unless and until we can
8	conclude with respect to particular types of
9	behavior that there is another type of test that we
10	have in predatory pricing that more specifically
11	advances the balance of maximizing consumer welfare
12	for that particular type of conduct.
13	The second thing that I would like to see
14	come out of the report, and this may be a little bit
15	outside the direct question of the adoption of
16	substantive rules under Section 2, is I think that
17	there are two powerful ways in which our analysis of
18	Section 2 substantive standards gets distorted by
19	things that don't directly relate to the merits of
20	Section 2 liability, which is, first, the prospect
21	of treble damages in private litigation, and the
22	second is the question of the scope of privileges
23	and immunities.
24	I think just as in our analysis of patent
25	reform, I think many people in the antitrust

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

25

1	community thought it is important not to remedy
2	problems with the patent system by adjusting
3	antitrust.
4	In the same way, I think it would be
5	important to try not to distort our analysis of
6	substantive antitrust analysis because of the fear
7	of treble damage liability, and if there is a
8	perspective that that is influencing or has a
9	powerful negative effect in terms of how Section 2
10	is being applied, that the agencies I would
11	encourage to address that head on as something that
12	Congress needs to address.
13	And in the same way, on sort of the opposite
14	side, I think that the ever-expanding scope of
15	privileges and immunities, the ability of people to
16	protect conduct that otherwise would be subject to
17	Section 2 is probably the single biggest deterrent
18	to the ability of the agencies effectively to
19	enforce against anticompetitive conduct.
20	That also would be an issue for the agencies
21	to identify for Congress and for the courts.
22	MR. BARNETT: Not hearing a lot of support
23	for a single unified test.
24	If I can turn to a slightly more specific
25	question, I guess, which is do you think that there

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

26

1	should be particular safe harbors, maybe conduct
2	specific or conduct-specific safe harbors under
3	Section 2, and if so, what are a couple of the areas
4	you would focus on?
5	I don't know if -- Greg or Jeff, you haven't
6	jumped in yet. If you want to tackle that one
7	initially.
8	MR. EISENACH: Let may say two things.
9	First of all, in my view, we have missed the
10	biggest issue in the room, and it is not in the
11	room, it is a couple thousand miles away across the
12	Atlantic and across the Pacific.
13	I agree with Jim, the LePage's decision
14	was -- what does Obi-Wan Kenobi say -- a powerful
15	disturbance in the force, and we all felt that
16	something bad had happened.
17	But that was a perturbance in a vastly more
18	settled pond than what we see going on around the
19	world.
20	I think reading the Article 82 Green paper
21	is in many ways an exercise in cognitive dissonance
22	for American antitrust professionals.
23	I guess if I were to suggest a number one
24	priority, both from a substantive perspective and
25	from the procedural perspective of venue shopping

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

27

1	and so forth, one of them has got to be trying to
2	continue the process of achieving convergence in the
3	major antitrust venues around the world. The EU is
4	not alone.
5	So I didn't want to let that go.
6	The second thing is that it seems to me that
7	the dichotomy between safe harbors and presumptions
8	on the one hand and a complete consumer welfare
9	approach on the other hand is a false one, and I
10	think it is captured in Doug's comment.
11	The question that Doug leaves me with is
12	what is the underlying analytical basis of the rules
13	that we do adopt? If it is not a consumer welfare
14	standard, then I don't know what it is.
15	I think our current safe harbors are quite
16	unsophisticated ones in many cases. I find it
17	inexplicable that 40 years after we began departing
18	from the structure conduct performance paradigm, we
19	are back at a point where the share of the number
20	one firm is somehow the proposed safe harbor in the
21	first step of a market power test.
22	I don't know what 75 percent or 50 percent
23	or 40 percent means out of context. And surely we
24	can state the safe harbors in more sophisticated
25	ways.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

28

1	But it does not seem to me that there is any
2	necessary conflict between a safe harbor test or a
3	series of safe harbors or presumptions on one hand
4	and a consumer welfare analysis on the other hand.
5	Had Microsoft had some legitimate business
6	purposes for some of the conduct for which it was
7	found liable in the Court of Appeals ruling, it
8	might not have been found liable.
9	That's a good example, I think, of a
10	presumption for a safe harbor which very much is
11	within the context of the whole rule of reason
12	analysis.
13	CHAIRMAN MAJORAS: Can I just follow-up on
14	that for a second?
15	I would like to see what others think about
16	that.
17	When we look at what the Court of Appeals
18	did in Microsoft and we talk about it as a balancing
19	test, I have always looked at it as a weighted
20	balancing test.
21	I think we are right about this. If you
22	read, as the Court of Appeals went through every
23	allegation of conduct, any time Microsoft put up any
24	plausible business justification for it, that ruled
25	the day and that was the end of it.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

29

1	It was just, I think when Microsoft said
2	"no, actually we didn't do those things," that then
3	the court said "oh, yes, you did, and because you
4	said you didn't, you didn't put forth a
5	justification, therefore you lose on that one."
6	It seemed to me the balancing test was
7	pretty weighted.
8	What do people think about that? Does that
9	make you feel better or worse about if the so-called
10	balancing test ended up sort of dominating in this
11	area going forward?
12	I know Doug is dying to weigh in.
13	MR. MELAMED: I think you are completely
14	right that the Microsoft Court never in fact
15	balanced.
16	In the two instances I believe it found that
17	there was a legitimate justification, and that was
18	the end of the analysis. Microsoft won.
19	In other instances, either because Microsoft
20	didn't advance a justification or the court rejected
21	it on the facts, Microsoft lost.
22	Let me comment on this idea of balancing
23	rule of reason in Section 2. It is a meaningless
24	concept. It is at best a throwback to the Chicago
25	Board of Trade case.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

30

1	In collusion cases, we know that rule of
2	reason means, did the agreement increase or decrease
3	the outcome of the parties to the agreement.
4	There is no metric, no meaning to rule of
5	reason, where you have both benefits and harms and
6	you are trying to balance them or, in Hovenkamp's
7	terms, assess proportionality.
8	As to safe harbors, I agree with Rick.
9	There ought to be a safe harbor where the conduct
10	did not exclude rivals or create or maintain
11	monopoly power.
12	And on the other extreme, I think that cheap
13	exclusion and other forms of naked exclusion, in
14	which there is no efficiency you can condemn the
15	conduct if it excludes rivals and injures
16	competition, without more.
17	But to talk about rule of reason or
18	balancing as a solution to the problem where you
19	have both benefit and harm it seems to me is
20	nonsense. And I don't think any court does it.
21	My experience is that courts find either a
22	justification, in which case defendant wins, or no
23	justification, in which case plaintiff wins.
24	It seems to me talking about rule of reason
25	is an empty vessel that leads courts to do what the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

31

1	LePage's court did, which is to say "I don't know
2	how to balance this, I don't know what to do with
3	this" and then come up with some crummy law because
4	it finds no guidance in the prior cases.
5	MR. PITOFSKY: The balancing test is the
6	baseline of all of antitrust.
7	The rule of reason compares procompetitive
8	justifications, anticompetitive effects.
9	Is there another way to get there without
10	examining the anticompetitive effects? That is true
11	of exclusive dealing, true of tying, true of
12	virtually everything regulated by antitrust, joint
13	ventures.
14	Merger is really a rule of reason analysis.
15	Why do you single out Section 2 of the Sherman Act
16	as an area where balancing is nonsense?
17	MR. MELAMED: Because I think of it as
18	collusion versus exclusion, not Section 1 versus
19	Section 2.
20	If you and I agree to a joint venture, we
21	can ask a simple question. Do the efficiencies
22	trump the market power? That is, does our output go
23	up or down?
24	If you exclude me from the market because
25	you have a more efficient exclusive dealing

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

32

1	agreement that enhances your ability to distribute
2	your product, you have the efficiency gains to you
3	and the exclusion to me and the consequences for my
4	customers.
5	I don't know of an algorithm that makes any
6	sense for weighing those two against each other.
7	MR. BARNETT: Rick.
8	MR. RULE: The only point I would make is
9	that, in this case, you are both right, I would say.
10	Bob's observation is sort of fundamentally
11	true about antitrust. Inherently in antitrust, you
12	are trying to balance harms to consumer welfare
13	against gains to consumer welfare.
14	I think Doug is right in the sense that it
15	becomes infinitely more difficult to make that
16	operational in a Section 2 context for a variety of
17	reasons.
18	So I agree with Doug that there is a need in
19	light of that to look for, if you will, operational
20	rules that incorporate that sort of insight of
21	balancing, but it is done in a way that courts can
22	actually manage.
23	You could argue that maybe they didn't do
24	such a great job in the Microsoft case. My
25	perspective is a little different than Debbie's, for

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

33

1	perhaps obvious reasons.
2	I think a lot of the company's
3	justifications were given the back of the hand,
4	frankly.
5	But I do believe that -- and I think this is
6	pretty consistent in Section 2 -- there is this
7	tendency, although it is a very difficult hurdle for
8	defendants to get over, but if defendants can show
9	that their conduct has a legitimate justification
10	for it, it typically is a good defense to a Section
11	2 claim, regardless of its impact.
12	I think that is probably an appropriate way
13	to approach it. Maybe Doug agrees with that.
14	The concern I have always had with a lot of
15	these tests is that at the end of the day, you have
16	to conclude that the conduct actually does exclude
17	somebody.
18	One of the reasons that you look at the
19	number one firm's market power, I would say, is a
20	legal reason. Section 2 talks about monopolization,
21	for better or worse.
22	That concept, other than a firm's market
23	power and its position relative to its competitors,
24	is meaningless. You have to give some meaning to
25	the law. That is what the law is.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

34

1	That's the single basis for attacking
2	unilateral behavior.
3	MR. PITOFSKY: The sentence was there are a
4	number of reasons why the rule of reason works in
5	many areas of antitrust but not Section 2.
6	I would be curious as to what those other
7	reasons are.
8	MR. RULE: If I said that, I'm not sure --
9	I think the concept of reasonableness is the
10	appropriate way to approach it.
11	The question of what the rule looks like in
12	Section 2 is more difficult.
13	One, it is more difficult because, unlike
14	Section 1 where you have an obvious target which is
15	an agreement that is in some way explicit between
16	two parties and you can look at it, in Section 2,
17	the conduct is not that explicit. It tends to be
18	implicit. It is something a company has done
19	unilaterally.
20	It is also very difficult to extricate it
21	from all the other competitive conduct that a
22	company engages in and evaluate it that way.
23	You have the fact that intent evidence, in
24	my opinion, is completely worthless in this area,
25	because you can't distinguish intent evidence that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

35

1	shows a desire to be vigorously procompetitive or
2	vigorously anticompetitive.
3	You also have the fact that -- and this was
4	really Doug's point, which was perhaps his principal
5	point -- unlike Section 1, where you can look and
6	say, "okay, gee, we have an agreement and what does
7	it do to market power, does it create it, is it an
8	exercise of market power?"
9	In Section 2, it is always indirect. First
10	off, we don't condemn a company unilaterally from
11	exercising market power.
12	One of the things that's interesting about
13	Trinko is the point the court makes that, rather
14	than condemning a monopolist for charging monopoly
15	price, we actually want him to do that because
16	that's his reward if he has gotten it through luck,
17	skill or foresight in doing it.
18	So instead, in a monopolization case, what
19	you are looking at is some sort of indirect impact
20	because there is an adverse effect on a competitor,
21	which you then have to translate into some impact on
22	consumer welfare.
23	Then you have to compare it with the
24	procompetitive benefits. That's very difficult.
25	That goes sort of to Doug's point.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

36

1	There is no algorithm for making that
2	comparison that I'm aware of from economists.
3	Instead, you have to try to develop rules,
4	whether they are safe harbors, whether they are sort
5	of general market power screens or something,
6	because I think saying that you are going to
7	directly measure and balance the procompetitive and
8	anticompetitive effects is probably fooling yourself
9	and the courts because it is not really possible.
10	Instead, you have to come up with rules that
11	are directed to trying to make that balance but
12	probably in some kind of gross fashion.
13	CHAIRMAN MAJORAS: I have a question about
14	the safe harbor concept.
15	Before I do, Greg, you have been so patient
16	down there. Is there anything you want to add on
17	any of these topics?
18	MR. SIDAK: I was going to go off in a
19	completely different direction.
20	Okay. I think that one of the big questions
21	that Section 2 poses is whether the jurisprudence in
22	this area is robust with respect to alternative
23	objective functions of the firm, alternative revenue
24	models, alternative production technologies.
25	By that, I mean suppose you change the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

37

1	assumption that a firm is a profit maximizer. Does
2	our existing jurisprudence on predatory pricing, for
3	example, give us much guidance?
4	It is not such a hypothetical question. For
5	example, the U.S. Postal Service is now subject to
6	antitrust -- it has had its antitrust immunity
7	lifted with respect to products that are not within
8	the statutory monopoly.
9	The last time I checked, the U.S. Postal
10	Service was not a profit maximizer.
11	With respect to revenue models, implicit in
12	a lot of the discussion we have had so far is that
13	we are talking about product markets that are pretty
14	easy to get our arms around, relatively mature
15	products.
16	What if we are talking about some of the
17	kinds of products and services that are at the
18	intersection of the Internet, telecommunications,
19	financial services and the like, where you have
20	multisited markets, you have multiproduct firms.
21	We can all agree that consumer welfare is
22	what we are trying to maximize. But which
23	consumers?
24	A given business practice may result in some
25	service being given away for free to one set of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

38

1	consumers. And that clearly benefits them. But is
2	there an adverse effect on some other set of
3	consumers?
4	So I think the consumer welfare objective is
5	just the beginning of the analysis.
6	When we are looking at some of these more
7	complex markets with multiple sides or firms that
8	are multiproduct firms, in which they may be
9	subsidizing a particular product in order to
10	stimulate the network effects and then with respect
11	to the production technology point, I think that
12	antitrust jurisprudence, compared to the traditional
13	law and economics of sector-specific regulation is
14	not very agile with respect to multiproduct firms.
15	I think this is one place where the
16	Europeans actually have shown some greater skill
17	than American courts.
18	In a case like the Deutsche predatory
19	pricing case in the EC, where they explicitly
20	recognized the multiproduct nature of the firm and
21	had to calibrate the predatory pricing rule to
22	reflect the fact that there were multiple products
23	involved.
24	So they used Jerry Fowell-Haber's
25	combinatorial cost test to try to establish what the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

39

1	appropriate price floor was for the particular
2	service in question that was allegedly being priced
3	below its cost.
4	So I think that the robustness of Section 2
5	jurisprudence across these different economic
6	dimensions is an important issue.
7	The other really big thing -- and I will
8	stop here -- is remedies and evaluation of the
9	efficacy of enforcement and of particular remedies.
10	We don't have much of a tradition. I'm not
11	sure we have much of a tool kit for knowing whether
12	we are systematically improving or reducing consumer
13	welfare over the long haul.
14	Much of the discussion about whether one
15	kind of rule is better than a different kind of rule
16	is really a question of are we minimizing the sum of
17	type 1 and type 2 errors under one approach rather
18	than another.
19	I don't know how we can possibly answer that
20	question unless we have some sort of time series to
21	look at.
22	Lawyers, that's not their stock in trade to
23	do that sort of thing. It is a very difficult task
24	to undertake.
25	CHAIRMAN MAJORAS: I agree with you on

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

40

1	remedies. I'm looking forward to discussing that
2	further with you.
3	I know Jim Rill was going to make a comment.
4	MR. RILL: I would just as soon follow-up if
5	you are going to start on safe harbors. If you want
6	to lead that off.
7	CHAIRMAN MAJORAS: Yes, I will. What is
8	interesting is, Jeff, I understand your point about,
9	for heaven's sakes, when you talk about safe
10	harbors, aren't you really talking about a market
11	share of safe harbor, and then aren't we going
12	backward, not forward, in terms of structural
13	analysis.
14	I heard what Doug said in agreeing with Rick
15	on what the safe harbor ought to be. That requires
16	some real analysis to get there.
17	A safe harbor not based on structural
18	presumptions might help you if you are actually in
19	court because it gives you a better chance of
20	winning.
21	How does that help lawyers who are
22	counseling their clients and trying to keep them out
23	of there initially?
24	What kind of a safe harbor can we have that
25	is truly meaningful and keeps people out of the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

41

1	legal system from the very beginning?
2	MR. EISENACH: Just very briefly. There are
3	others wiser than me on this.
4	First, I'm not opposed in any way to a 75
5	percent safe harbor or a 70 percent safe harbor. It
6	is better than a 50 percent safe harbor.
7	My point really went to the notion that
8	surely we can do better than share of the top firm
9	as a metric. That surely can't be the best we can
10	do.
11	But the second point would be that, again, I
12	think that the metrics can become more robust and
13	more sophisticated without becoming less useful.
14	Also, do we have it upside down when we look
15	at market shares first and entry second? I think we
16	do.
17	CHAIRMAN MAJORAS: Interesting. Jim?
18	MR. RILL: I think history has embedded us
19	with the notion of at least a market share test for
20	a safe harbor, at least as a starting point, only as
21	a starting point.
22	The International Competition Network
23	recently surveyed, as part of its single firm
24	conduct working group, the question of whether or
25	not -- first of all, I think something like 70

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

42

1	percent agreed that consumer welfare of the 35
2	nations that responded to the questionnaire, that
3	consumer welfare was the appropriate underlying
4	fundamental principle of monopolization Section 2,
5	Article 82 and related enforcement technology
6	techniques. But very little probing beyond that as
7	to what consumer welfare meant.
8	I think I have to say that Bob is a little
9	bit simplistic on this notion, and I think there is
10	a lot more latitude, but that is another issue.
11	I think that is a starting point. Again,
12	any number, about 70, 80, 90 percent of respondents
13	to the questionnaire would use a safe harbor
14	threshold of some level of market share, market
15	power, if you will.
16	Now, some of those safe harbors are rather
17	low. I think Japan is around 10 percent, which
18	doesn't give me a lot of comfort. 70 percent sounds
19	reasonable to me, maybe a little higher.
20	But I think we can get beyond that. I think
21	there is enough -- a lawyer quite clearly can
22	demonstrate, an economist can demonstrate that there
23	is a rich body of law in the United States stemming
24	from the law of predatory pricing which can bring
25	into the notion of consumer welfare certain

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

43

1	operational tests, if you will, that can be safe
2	harbors applicable not only in the predatory pricing
3	area but with some further depth analysis into areas
4	that go beyond single firm predatory pricing, in
5	fact, in all pricing areas, bundled pricing, loyalty
6	discounts and maybe developing into the areas of
7	coercive tying, one wants to think about not
8	contractual tie but price-related tie.
9	I think a thought given to that kind of an
10	operational safe harbor approach is not inconsistent
11	either with the unilateral or unitary test.
12	It doesn't seem inconsistent with a consumer
13	welfare analysis stemming from some of the
14	literature, at least in the Trinko decision and more
15	recently in the Weyerhaeuser decision, where the
16	Supreme Court provided that kind of approach to a
17	safe harbor from a legal operational basis and would
18	provide significantly greater clarity to those of us
19	who are trying to counsel companies and to
20	enforcement agencies as they move to the next stage.
21	MR. BARNETT: I think Jim's comments
22	actually began to quite conveniently and
23	appropriately blend into our next topic, having to
24	do with a definition of what is monopoly power and
25	by your reference to defining that through market

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

44

1	shares.
2	Debbie, I think you are going to lead off on
3	that.
4	CHAIRMAN MAJORAS: I will.
5	Doug, you have been dying to jump in on this
6	issue. I think it relates. If you want to go
7	first.
8	MR. MELAMED: I will be very brief.
9	Debbie, I was very glad that you asked the
10	safe harbor question in terms of the impact on
11	counseling rather than just the impact on
12	litigators, because the impact of antitrust rules in
13	litigation, it seems to me is much less important
14	than the impact of those rules on the millions of
15	decisions that businesses make every day that don't
16	reach the courts, that is, on the guidance that
17	antitrust law gives to the business community.
18	From my experience in counseling, market
19	share-type screens are of limited value because
20	market share depends on market definition, and it is
21	a binary concept and we are often sitting there,
22	saying well, gidgets might be in the market with
23	widgets, but they might not be and who knows.
24	In my experience, much more useful to the
25	client are guidelines and safe harbors that focus on

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

45

1	the nature of the defendant's conduct, things like
2	is the price below your avoidable cost, does it make
3	business sense, are you sacrificing a profit,
4	whatever it may be.
5	Even rules of that type I think are bad
6	rules are useful for counseling -- rules such as:
7	Is the exclusive dealing contract for a duration of
8	a year or less?
9	Those things that enable the defendant to
10	look at his conduct are much more valuable as safe
11	harbors than those that require him to analyze the
12	market.
13	CHAIRMAN MAJORAS: Okay.
14	Susan, as we look at the concept of monopoly
15	power and we typically begin the analysis with that
16	in a Section 2 context as well as in a Section 1
17	context, I should say -- welcome, Tim.
18	MR. MURIS: Thanks.
19	CHAIRMAN MAJORAS: As we look at this, do
20	you think it is useful for us to establish a sort of
21	conclusive presumption on market share?
22	We have had a couple comments here that the
23	market share screens are really not that useful and
24	you have to do so much analysis anyway in order to
25	define the market that it is not that useful.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

46

1	You have certainly been on the enforcement
2	side. What do you think about those kinds of safe
3	harbors?
4	MS. CREIGHTON: I think both Professor
5	Elhauge and also maybe Tom Krattenmaker and
6	Professors Lande and Salgo have written a couple of
7	articles talking about how market power -- not
8	market power -- the percentage of the market that
9	you control actually can be helpful as direct
10	evidence regarding how profitable is it likely to be
11	to you and both your incentives and your ability to
12	enter into some kind of exclusionary conduct.
13	So it can be direct evidence and quite
14	important in that way.
15	I do get concerned about using, at least in
16	attempt cases, as a screen, because I think if you
17	looked at Unocal or Rambus, for example, without
18	getting into the -- sort of any standard-setting
19	case, the person may have had no market share at all
20	in whatever the relevant market was.
21	That does not necessarily dictate how
22	likely -- what the market share would have been or
23	their market power would have been if the
24	exclusionary conduct was successful.
25	So I would be concerned about saying it is

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

47

1	always required as a preliminary step before you get
2	to the question of -- one of the advantages that I
3	think or one of the things that American law
4	emphasizes which maybe the Europeans don't as much
5	is I think for them, they really do focus on market
6	share dominance, and then they have very strict
7	definitions of if you are one of those folks, what
8	can you do.
9	In the course of that, they really lose
10	sight of the question of the causation and whether
11	or not the conduct is conduct that we are concerned
12	about in terms of increasing barriers to entry or
13	otherwise increasing somebody's market power in a
14	way we would be concerned about.
15	I would be concerned also about using a
16	market power screen in the first instance to make
17	sure we don't lose sight of that important
18	additional causation requirement.
19	I think that could be a danger.
20	CHAIRMAN MAJORAS: On the question of
21	durability, I know that in prior panels the
22	panelists really agreed that we need to look at
23	market power and whether it is both substantial and
24	durable.
25	Susan, you certainly but I think everybody

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

48

1	today now does so much work in dynamic industries
2	and technology industries in which even if you have
3	market power, it might be quite fleeting. There may
4	not be a durability.
5	Does that make it even less the case today,
6	that we should be looking first at market share
7	screens as a way to at least start to get into the
8	analysis?
9	Bob?
10	MR. PITOFSKY: I think you put it just right
11	toward the end of your remarks.
12	Marketshare is the ramp that leads you into
13	the analysis. The problem is sometimes judges and
14	lawyers think the ball game is over because of the
15	way in which the market has been defined. We
16	shouldn't do that.
17	When you get to the end of the analysis and
18	you look at conduct and barriers and all that, you
19	go back and see if your market share analysis is
20	correct in light of all these factors.
21	Of course, substantiality and durability are
22	critical. If you have market power, but it only
23	survives for a year and then is displaced by some
24	other product that is not really market power.
25	We know the barrier to entry is important.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

49

1	This applies to high-tech. I have always been an
2	admirer of Andrew Groves' book "Only the Paranoid
3	Survive."
4	The whole idea of Learned Hand that market
5	power is a narcotic and competition is a stimulant,
6	you can't say that about these big high-tech
7	companies. They are extremely aggressive in their
8	innovation, and that's a factor that has to be taken
9	into account.
10	But unless you start with market power, I
11	don't know where else you start. It gets you going,
12	because some things, some behavior engaged in by a
13	company with 10 percent of the market is legal and
14	is illegal if the firm has 90 percent of the market
15	is illegal.
16	You have to address that question at an
17	early point. I skipped over the safe harbor.
18	Let me just say that first of all, I'm not
19	comfortable with safe harbors. I like rebuttable
20	presumptions because there are too many quirky
21	situations.
22	Somebody has 40 percent of the market but
23	everybody else has one percent each. So I think
24	that presumption of a safe harbor is rebuttable.
25	Secondly, the safe harbor is going to vary

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

50

1	according to the behavior you are dealing with. We
2	have safe harbors for exclusive dealing.
3	We have safe harbors for tie-in sales in
4	terms of the market power of the seller instituting
5	that program, 30, 40, 50 percent and so forth.
6	When you get to lying to the Patent Office,
7	I don't think there is a safe harbor. I don't think
8	there should be a safe harbor.
9	So I think safe harbors, of course, are
10	useful to people who are advising firms about what
11	they can and cannot do, but they should vary
12	according to the nature of the conduct.
13	MR. BARNETT: What if you lie to the Patent
14	Office and get a patent that actually confers no
15	market power, what do you mean there is no safe
16	harbor? Have you violated Section 2 then?
17	MR. PITOFSKY: If you lie to the Patent
18	Office? You are talking about Walker Process
19	insisting on defining the relevant market in order
20	to make out a violation for lying to the Patent
21	Office?
22	MR. BARNETT: The statement was if you lie
23	to the Patent Office, there should be no safe
24	harbor.
25	I'm just wondering what that means in terms

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

51

1	of if you get a patent, I think most of us agree
2	that doesn't necessarily give you market power if
3	you end up with a patent which does not give you
4	market power.
5	Have you violated Section 2 or not?
6	MR. PITOFSKY: Fair enough. My answer is
7	there are no redeeming virtues to lying to the
8	Patent Office, none whatsoever.
9	MR. BARNETT: I understand. But if I can
10	perhaps -- I thought it was a yes or no question.
11	MR. PITOFSKY: Okay. Here's my answer to
12	that.
13	CHAIRMAN MAJORAS: You are back in Congress.
14	MR. PITOFSKY: Horizontal price fixing may
15	confer no market power. We declare it illegal.
16	I think lying to the Patent Office is the
17	same thing.
18	MR. BARNETT: Fair enough.
19	CHAIRMAN MAJORAS: We have talked about --
20	Bob, you and some others have said if we don't start
21	with market share, where do we start. We have
22	started there for very long time.
23	But Jeff Eisenach said why don't we think
24	about entry first. I think that's what you said,
25	Jeff.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

52

1	Anybody have any -- Greg?
2	MR. SIDAK: I haven't heard anybody utter
3	the words price elasticity. That's what I care
4	about. I don't care about market shares or entry.
5	If I can directly observe the price
6	elasticity of demand, I can make an inference about
7	whether it is profitable or not profitable to raise
8	price.
9	Let me give you a hypothetical example.
10	Suppose some high-tech industry, a firm has 40
11	percent of the market, casually defined.
12	It raises the price by 10 percent, and its
13	competitors over the same period of time lose market
14	share.
15	Would we infer that there is not a problem
16	because the market share is only 40 percent and that
17	is way below Judge Hand's ALCOA threshold or would
18	we look at a price increase or loss of competitor
19	market share and say that is a more direct set of
20	facts that elucidates what the price elasticity of
21	demand is?
22	CHAIRMAN MAJORAS: Rick Rule, could you
23	counsel a client on that basis?
24	MR. RULE: On price elasticities?
25	CHAIRMAN MAJORAS: Not you personally. I

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

53

1	have no doubt you could.
2	MR. RULE: I have generally not had to do
3	that, fortunately.
4	But there are always difficulties, and you
5	have to exercise care when you are counseling a
6	client. But frankly, I have always found the market
7	share requirements of Section 2 to be helpful in
8	terms of advising clients.
9	There are edge cases where it can be a
10	little difficult, and you can tell the client, "gee,
11	I know you don't think you have a monopoly and that
12	you are in a very competitive world, but there are
13	ways in which a court could find the opposite, so
14	you have to exercise some care."
15	But for a lot of companies, given the nature
16	of the industries they are in and what they are
17	doing, it is pretty clear that they don't have
18	market power, and you can worry about other parts of
19	the antitrust laws.
20	I will say that it is probably more
21	difficult as technology has moved along and as the
22	economy has gotten somewhat more dynamic and
23	complex, particularly for information industries.
24	It becomes a little more difficult to use the market
25	power and monopoly power market share screen that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

54

1	traditionally we have used.
2	And I guess for that reason, when you are
3	counseling clients, you kind of have to have in the
4	back of your mind that there could be a way to
5	define the market that would suggest they do have
6	monopoly power.
7	So then you go directly to conduct. And in
8	those industries, particularly, conduct safe harbors
9	would probably be very helpful.
10	So to some extent, I think conduct safe
11	harbors are appropriate there. I will also say,
12	interestingly, in information industries, you rarely
13	get that concerned, at least I do, about pricing
14	issues. Because if you think about it, if they are
15	information industries, generally marginal cost will
16	be pretty low and you will recognize that predatory
17	pricing issues are not that problematic.
18	Generally, I think market share screens have
19	worked. They are more complex today, but they have
20	some value in counsel.
21	MR. BARNETT: Related to that, if I could
22	follow-up with Greg for a little bit, the economists
23	generally tell me that if you think about perfect
24	competition, the way you deal with that is you graph
25	that and it would be a perfectly horizontal demand

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

55

1	curve, and if you have downward sloping demand
2	curve, you have some degree of market power.
3	And if you measure that directly, it is
4	probably true that the vast majority of firms in the
5	United States have a somewhat downward-sloping
6	demand curve.
7	Does that mean they all have market power
8	and we should just move on from there? Or should we
9	try to deal with that in some meaningful sense to
10	help in part from a counseling perspective?
11	MR. SIDAK: Of course, they may have
12	differentiated products that explain the downward
13	slope of their firm demand curves.
14	The slope of the demand curve, of course,
15	doesn't tell you whether the firm is earning
16	monopoly rent or just quasi, a risk-adjusted return
17	on investment in innovative activities, for example.
18	So I don't think that the downward-sloping
19	demand curve itself is a cause for antitrust
20	intervention.
21	In terms of the market share, market power
22	filter that we have been discussing, I think it is
23	possible to directly infer something about the price
24	elasticity of demand for a firm even in the absence
25	of market shares if you have certain evidence.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

56

1	So in other words, I don't think you should
2	necessarily back away and say, well, this is way
3	below Judge Hand's threshold in ALCOA, there is no
4	way this could be a monopoly problem. It might be a
5	monopoly problem.
6	CHAIRMAN MAJORAS: Tim, you had a comment.
7	MR. MURIS: I thought Tom's point was quite
8	perceptive. It is not just differentiated products.
9	If you walk on the Mall, any hot dog vendor
10	who raises his price won't lose all his sales. That
11	means the demand is a downward-sloping curve. The
12	reason is transaction costs more than anything else;
13	in a world of positive costs, just about everybody
14	has a downward-sloping demand curve.
15	This fact has profound implications for
16	antitrust economics. Ben Klein has written the best
17	about this in his analysis of the Kodak case and
18	other articles.
19	It means that it is difficult to have simple
20	uses of Lerner indexes and downward sloping demand
21	as measures of anything meaningful.
22	CHAIRMAN MAJORAS: Any comment? No?
23	MR. SIDAK: A common problem when you start
24	looking at industries that are subject to some kind
25	of public service regulation, of course, is that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

57

1	they may be compelled to sell products at low prices
2	or even below costs.
3	So the Lerner index actually has its
4	causation reversed. They have a high market share
5	because they are compelled to charge low margins or
6	negative margin.
7	I agree with Tim that the Lerner index is
8	uninformative and potentially misleading in
9	situations where you have significant economies of
10	scale.
11	MR. BARNETT: Jim, I will turn to you for
12	our next topic to lead off, because that is bundled
13	discounts. You have already revealed a particular
14	interest in that area.
15	We recently had a report issued by the
16	Antitrust Modernization Commission that addressed
17	this topic and set forth a three-part test to
18	determine whether or not there is a violation of
19	Section 2 from bundled discounts.
20	Just briefly, the first prong is allocating
21	all of the discounts to the competitive product --
22	sometimes referred to as the Ortho test -- second,
23	whether or not the defendant -- whether it is below
24	cost under that measure. Second, whether or not the
25	defendant is likely to recoup those losses. And

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

58

1	thirdly, whether the bundled or rebate program has
2	had or is likely to have an adverse effect on
3	competition.
4	Aside from the fact that the third prong
5	seems to sort of ask the ultimate question there,
6	the question is is this appropriate standard, is it
7	appropriate as a safe harbor but perhaps not the
8	standard or is it just something we should be
9	looking in a different direction?
10	MR. RULE: First of all, I think the AMC is
11	looking at it only when it relates to conduct by
12	someone who is judged to be a monopolist.
13	Moving on from that to the operational test,
14	I have some difficulty with let's call it the Ortho
15	or AMC allocation formula, both from an operational
16	and from, I think, an analytical standpoint.
17	From an operational standpoint, the
18	allocation itself of the totality of the discount
19	across to the single let's call it target product
20	creates something of a daunting task, and there is a
21	margin or opportunity for error there that I think
22	is quite substantial.
23	Secondly, from an analytical standpoint, I
24	think maybe it is operational as well, it raises the
25	problem of double counting or multiple penalties.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

59

1	Just to take a hypothetical industry, if
2	someone sues on Post-its, and someone else sues on
3	sponges, and someone else sues on tape, and there
4	are three cases going on at the same time, does one
5	allocate the totality of the package discount to
6	each of those products, and in what position does
7	that put the defendant in?
8	However, I think I prefer that there is a
9	solid operational test to safe harbor. The
10	proposition that I think is embraced in Tim's
11	statement to the AMC is that the allocation of total
12	cost to total bundles would be a better way of
13	looking at a test that might suggest illegality.
14	On the other hand, it is possible that the
15	allocation test or the Ortho or AMC formula of
16	allocation, would be appropriate as a safe harbor.
17	This is the position taken in the brief of
18	several law professors recently filed in the Ninth
19	Circuit in the Peace Health case, Professor Crane
20	and others.
21	Recognizing that the difficulties that we
22	have suggested with that test as a presumption of
23	illegality, it might serve a purpose at least of a
24	safe harbor if practicable.
25	So far as recoupment is concerned, I think

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

60

1	in a pricing case, recoupment should be an element
2	of the offense and should be considered as part of
3	the potential safe harbor.
4	MR. BARNETT: Bob?
5	MR. PITOFSKY: Let me second what Jim has
6	just said and then let people take shots at it.
7	First of all, back to Doug's excellent point
8	in opening this whole discussion.
9	What is this all about? What are we
10	quarreling about here? It seems to me the point of
11	bundled discounts is it gives consumers a break. We
12	ought to not be too aggressive in deterring it.
13	We should not overdeter it, it, but be
14	careful in this area. Second, it seems to me to be
15	more sensible, as the Aveeda-Turner Treatise
16	originally said about this question when it first
17	came up, you want to allocate the discounts product
18	by product rather than put all of the discounts to
19	one product.
20	There is a serious danger that will drive
21	the price of that product below whatever predatory
22	pricing turns out to be.
23	So I share Jim's view. I think LePage's was
24	wrong, and if the court gets to overturn it and come
25	up with a more sensible rule, the better off we all

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

61

1	are.
2	MR. BARNETT: Tim?
3	MR. MURIS: Obviously, anything is better
4	than 3M, than turning it over to the jury.
5	The AMC deserves credit for trying to devise
6	a test. But there are serious theoretical,
7	empirical, and practical problems.
8	As Dennis Carlton said in the AMC report,
9	the bundled discounts can be used for procompetitive
10	reasons. For example, price discrimination can be
11	anticompetitive or procompetitive. It is difficult
12	to separate pro from anti and we need to be careful
13	for that reason.
14	The second theoretical problem is the
15	premise of the AMC allocation is to protect "equally
16	efficient competitors." The problem -- and there is
17	a nice footnote in the government's LePage's brief
18	about this -- is that someone who sells you one
19	thing that you want can't be as efficient as someone
20	who sells you two things that you want.
21	So the AMC's premise is a problem.
22	Moreover, empirically we know almost nothing that
23	tells us that there are anticompetitive problems
24	from bundling. Vernon Smith and I have put together
25	a paper that summarizes the work of his group, which

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

62

1	spent a lot of time using experimental economics to
2	take the theories of anticompetitive bundling and
3	show they actually hurt consumers.
4	Well, it was almost impossible to do. They
5	did find some ambiguous cases. Yet, if you do
6	anything to those ambiguous cases, bundling becomes
7	efficient. Thus, if the monopolist lacks a 100
8	percent share, if there are any efficiencies, like
9	transaction cost savings, and if you don't have very
10	strange-looking demand curves, bundling becomes
11	efficient. Obviously, experimental economics has
12	its limits, but it is certainly superior to simple
13	theoretical arguments.
14	There is also a tremendous practical
15	problem. Greg has done a lot of useful work in
16	valuing regulatory agencies, and there is some older
17	and good literature about allocating joint and
18	common costs. If you start trying to do this across
19	the products in a bundle, it is completely arbitrary
20	in terms of allocating these costs to some products
21	and not to others.
22	Finally, I do agree we need a safe harbor.
23	The Brooke Group allocation, the more general
24	allocation that Jim and Bob are discussing is the
25	one that I would support.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

63

1	CHAIRMAN MAJORAS: Doug?
2	MR. MELAMED: There is a lot of force to
3	Tim's points.
4	I completely agree that economies of scope
5	are relevant economies and should be taken into
6	account in the efficiency analysis.
7	I think there is a lot of force to Tim's
8	notion that maybe because we don't have a lot of
9	confidence that, bundling is likely over a lot of
10	cases to reduce consumer welfare, we should paint
11	with a broad brush and apply the Brooke Group test
12	to the package.
13	But, ultimately, I don't agree with Tim
14	because, first of all, I think the premise which Tim
15	didn't state but I think Bob did, that bundled
16	discounting is like single-firm price cutting --
17	that it is a price reduction that has short-term
18	benefits for the consumer -- is not necessarily
19	correct. In order to say that, we need to know what
20	the but-for pricing would have been. I think it may
21	well be the case that, in the absence of bundling,
22	the stand-alone prices would be lower than they
23	would be with the bundled offering provided. So the
24	discount might be mythic.
25	One can imagine situations in which one

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

64

1	would increase the price on the monopoly product and
2	use the margins there to subsidize below-cost
3	pricing on another product, and you can imagine some
4	competitive harm from that.
5	So where I come out is to think that the
6	AMC's three-part test -- ought to be a safe harbor,
7	but it shouldn't be the end of the analysis.
8	I agree with Dennis Carlton. I think his
9	articulation in the AMC Report is right. That's a
10	safe harbor. But you also have to -- Dennis
11	actually admitted this, although he is not a
12	supporter of the no economic sense test, he admitted
13	what he was articulating as his separate statement
14	was that no economic sense test.
15	You ought to allow the defendant and the
16	plaintiff to duke it out over whether the bundling
17	made economic sense.
18	MR. PITOFSKY: Very briefly.
19	MR. BARNETT: Sure.
20	MR. PITOFSKY: I have never seen a bundling
21	that you can have A, B, C separate price, if you
22	take all three, I will give you 10 percent off. I
23	have never seen a situation where that produces
24	higher prices than bundling produces.
25	More important, the idea that we should

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

65

1	somehow examine bundling by taking into account the
2	efficiency of the bundler and the efficiency of the
3	company that doesn't have the bundled offering, just
4	think about that from the point of view of
5	counseling.
6	Just think about the businessman saying,
7	"well, if I do this, will I be in trouble?"
8	"No, not if the other fellow is not equally
9	efficient as you and therefore is driven out. On
10	the other hand, if they are equally efficient and
11	this puts them out of business, you are in a lot of
12	trouble."
13	How does the businessman know what the level
14	of efficiency is? Not only doesn't he know his own
15	level of efficiency, but how is he possibly going to
16	know the level of efficiency of the other guy?
17	I think -- I have been there. I tried to
18	draft a subpoena to figure out whether the other
19	company was equally efficient. It was a disaster.
20	It wasted a lot of money and we never got anywhere.
21	MR. BARNETT: You are not going to get
22	private counselor subpoena power, I assume.
23	CHAIRMAN MAJORAS: I think Jeff wanted --
24	MR. EISENACH: I want to speak up in defense
25	of recoupment. And in the same spirit as earlier,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

66

1	speak about the importance of entry.
2	All of these behaviors are designed to
3	foreclose in the sense of capturing market share.
4	The question I think we want to look to is
5	whether enforcement offers a way of going forward to
6	police prices at or near the competitive level and
7	police behavior at or near the competitive level.
8	If recoupment isn't possible, then it seems unlikely
9	to me that enforcement is improving consumer
10	welfare.
11	MR. BARNETT: Can I ask, is there a
12	difference -- and maybe this would go to Jim and Bob
13	as much as anyone -- if the plaintiff comes in and
14	alleges a bundled discount, you apply the standard
15	that you were suggesting or the plaintiff comes in,
16	same set of facts, and says this is an illegal
17	tie-in.
18	Is it the same analysis? I assume we agree
19	that at some level a pricing structure could be
20	labeled a de facto tie-in and tying theoretically
21	could apply.
22	Does it matter what label the plaintiff puts
23	on it or is there some other way to distinguish
24	between those two types of claims?
25	MR. RILL: I assume you are talking about

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

67

1	what would be a pricing tie rather than a clear
2	contractual tie.
3	With respect to I think the unicorn of a
4	pricing tie, I see no reason why there would be any
5	different test as to what is the nature of the
6	plaintiff's claim.
7	I know that Hovenkamp and others would
8	suggest that tying analysis is the right analysis to
9	apply to bundled pricing.
10	At the same time, at the end of the day, he
11	comes out with a test that is very much like,
12	depending on when and what you read in Hovenkamp, it
13	is either Ortho or Brooke Group, depending on
14	whether it is the book or the most recent article.
15	I think the analytical formula should be
16	exactly the same. If it is time to apply tying
17	rules to Section 2, I think that's a good move, too.
18	The tying should be analyzed under Section 2
19	rather than as a per se offense as the courts at
20	least currently view it.
21	I see no reason why you would deviate from
22	the kind of safe harbor approach in tying as you
23	would in a claim that is a pure pricing claim.
24	MR. PITOFSKY: I must say that's a tough
25	one. The treatise position, as I recall it, is if

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

68

1	everybody takes the discount offer, it's a tie.
2	That doesn't mean it is illegal. It should
3	be treated as a tie.
4	If a relatively small number of people say
5	"I don't want that deal, I will stick with buying
6	separately," then you treat it generously. It is
7	not a tie; it is bundling. And for all the reasons
8	that we have already discussed here, it turns out
9	the customer gets a bargain.
10	That is about as generous as I think we
11	probably ought to go, although, as I say, I did
12	contend once that as long as you can buy the
13	products separately, if you can get them for less, I
14	wouldn't be unhappy if that were per se legal.
15	MR. RULE: I think the question about
16	tie-ins and comparing that to bundled discounts is a
17	good one because it points out one of the flaws in
18	the AMC rule and a lot of the rules, from my
19	perspective.
20	I think it is true that the kind of three
21	parts, at least the first part, ought to be viewed
22	as a safe harbor. And if that condition exists,
23	that you allocate all of the discount to the
24	supposed competitive product and the price is still
25	above some incremental cost, then it seems to me

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

69

1	that it ought to be in the safe harbor.
2	But ultimately the question of whether or
3	not it is illegal ought to be related to the degree
4	of exclusion or foreclosure that is created.
5	The problem in saying that that's not
6	incorporated in the AMC test is who knows what the
7	third step means. Maybe that's what they meant by
8	the third step.
9	I think noting that a bundled discount could
10	be viewed, under certain circumstances at least, as
11	a price tie points out, or to some extent
12	exclusionary conduct generally points out, the fact
13	that all of the tests ought to be focused at the end
14	of the day on the extent to which they exclude
15	competition, not just competitors from the
16	marketplace.
17	There ought to be some notion of that. For
18	example, if relatively few consumers actually take
19	the discount, then it is a little difficult to say
20	that there is some sort of exclusionary impact. And
21	that ought to be the end of the story, whether you
22	view it as a tie or bundled discount or anything
23	else.
24	One of the problems -- and this is one of
25	the problems I had with the unitary rules, profit

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

70

1	sacrifice and that sort of thing -- is they don't
2	focus on the degree of foreclosure or exclusion.
3	I think if you ignore that, you potentially
4	end up challenging a lot of conduct that is not
5	necessarily anticompetitive.
6	It is also the reason that I think the
7	incorporation of the recoupment test, as a couple
8	people have already said, in a number of different
9	areas is at least a start in terms of focusing on
10	exclusion, because the recoupment test sort of
11	presumes that there is exclusion and that there
12	cannot be reentry, and that's the way recoupment
13	occurs.
14	So at least the recoupment test has that
15	benefit. In my mind, at least, in predatory
16	pricing, that has been the principal innovation that
17	has made it less of a problem, because the cost
18	tests were always very hard and difficult and
19	time-consuming to litigate.
20	The recoupment test, which I think can
21	dispose of a large fraction of predatory pricing
22	cases and probably a lot of these other cases at the
23	end of the day, indicates that there is really no
24	harm to consumer welfare; there is no exclusion that
25	you need to be concerned about.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

71

1	MR. BARNETT: Tim?
2	MR. MURIS: Theoretically, tying is
3	different. The problem is in what is mostly the
4	vast wasteland of modern IO, of which I'm not a fan,
5	obviously, tying can be a problem.
6	What we know about bundling is that it is
7	efficient and the experimental evidence really
8	supports what Bob is saying. If it is really a
9	bundle, which means that it is not a tie, there are
10	people buying the bundle products as separate
11	products. The bundle thus is not a de facto tie.
12	It is hard for me to envision a case where
13	we would attack bundle. Yet from what we know about
14	the theoretical literature of tying and the lack of
15	evidence there is slightly more support for worrying
16	about tying.
17	There is a Sibley paper, which says that the
18	problem with bundling is that it is a de facto tie.
19	Yet, the second version showed you need to
20	have perfect competition to have a problem. Of
21	course, we don't have perfect competition.
22	So, the de facto tie didn't prove to be a
23	very strong reason to worry. We tried to test that
24	in the experimental setting. Again, that proved
25	something close to the empty set for anticompetitive

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

72

1	conduct.
2	MR. BARNETT: Doug?
3	MR. MELAMED: Two things. Rick repeatedly
4	said we ought to have a rigorous requirement of harm
5	to competition.
6	I assume we all agree with that. That's not
7	the issue.
8	Certainly at least one person who has
9	written in favor of a so-called unitary test -- I
10	think two of us actually did -- tried to make it
11	perfectly clear that of course you have to have
12	proof that the conduct had an impact, injured
13	competition, but then went on to say, let's talk
14	about a second way a defendant could win the case
15	even if the conduct excludes competition because a
16	better mousetrap could do that. Let's focus on the
17	conduct element.
18	I assume everybody agrees here we have to
19	have a rigorous competitive effects test.
20	On the question of, is it tying or is it
21	bundling and what is the difference, and listening
22	to Tim talk, I can't help but ask why are we
23	worrying about the kind of analogical issue of what
24	category does the conduct fall into.
25	To do that, we have to define the conduct.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

73

1	Then we have to put it into a category. Then think
2	of the rule for that category. We wind up with a
3	lot of formal distinctions and without overarching
4	principles to give guidance to a court like the
5	LePage's court when it has something that doesn't
6	fall into a specific category.
7	Why don't we simply think of the facts of a
8	case of bundling, for example, and ask, how do we
9	think we ought to analyze it, without worrying about
10	what is the better analogy -- predatory pricing or
11	tying or exclusive dealing or whatever the next
12	category of the day might be.
13	MR. BARNETT: If I can briefly follow-up
14	though.
15	If we abandoned the unitary test and are
16	going to apply different operational tests to
17	different contexts, doesn't that necessarily create
18	the need to decide which bucket you are in?
19	MR. MELAMED: I guess I would say we
20	shouldn't have that need.
21	CHAIRMAN MAJORAS: Really?
22	MR. RULE: Let me make one point.
23	It is nice when folks say that exclusion
24	ought to be an element. It wasn't really in the
25	government's brief, as I read it, when they

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

74

1	articulated the unitary effect test.
2	It is simply some harm to a competitor,
3	which is very different. That's what it said.
4	There was no quantitative exclusion.
5	The only place I have ever seen it is in
6	exclusive dealing cases. Even there, to some
7	extent, the government backed off of that in some of
8	the cases.
9	So you may be right and maybe that's a
10	standard. But that is not generally how it has been
11	articulated to the court.
12	If you look at what Judge Jackson said in
13	the Microsoft case in the District Court, that is
14	not how he viewed it.
15	The profit sacrifice test is generally
16	viewed as being a problem and negating the need to
17	actually look at whether there is a quantitative
18	measure of exclusion of competition from the
19	marketplace.
20	But if you are saying that, "no, in fact
21	that is a precursor and this is another way and all
22	the unitary test is designed to do is provide an
23	additional safe harbor," I guess I don't dislike it
24	as much as I thought I did.
25	But that's not the way I have ever seen it

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

75

1	articulated in any of the court's briefs and I
2	thought in your articles as well as others, but I
3	will have to go back and reread them.
4	CHAIRMAN MAJORAS: Anything else before we
5	move on to loyalty discounts?
6	I will ask a bridge question, bundled
7	discounts, bundled rebates and loyalty discounts.
8	And that is we do hear a lot that this is an area
9	within antitrust law in which everyone could use
10	more guidance. I certainly understand that.
11	But I have a question that's related which
12	is how big a problem is it that there isn't more
13	guidance? In other words, how often is this coming
14	up?
15	Obviously, you can't tell me in some
16	measured sense. I'm just curious, as you are
17	counseling clients, whether these are issues, these
18	pricing and discounting issues are sort of burning
19	on the agenda for clients on a pretty regular basis.
20	Doug?
21	MR. MELAMED: I think that, because there is
22	less, there is probably more confusion or unease
23	about the bundling law post LePage's, it is probably
24	an area where the clients and their counselors feel
25	a little less sure footed.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

76

1	It is a problem. Is it crippling the
2	American economy? No.
3	CHAIRMAN MAJORAS: I'm glad about that.
4	Jim?
5	MR. RILL: Look at some of the cases coming
6	up and you will see it is a problem.
7	You have cases that are for some strange
8	reason being focused in the Third Circuit on bundled
9	prices and loyalty discounts.
10	You have a case coming up in the Ninth
11	Circuit, Cascade, the Sixth Circuit, Wyatt, all of
12	which are being argued. And in the Ninth Circuit
13	District Court construction is literally lifted from
14	LePage's that resulted in a plaintiff's verdict
15	there.
16	Yes, it is an important problem.
17	Let me bridge, to use your term, to the
18	global aspect of the problem, because I think we
19	can't ignore and shouldn't ignore the uncertainty
20	and prevalence of the uncertainty surrounding these
21	kinds of practices overseas.
22	I think we are aware of circumstances in
23	Europe and the Far East where the law is, if you
24	will, less developed or developing, not developing
25	in the way we would want to develop it.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

77

1	I think to the extent that the views of the
2	United States in these areas could be made known and
3	enforcement agencies in these areas could be made
4	known explicitly in an effort through international
5	organizations to secure convergence, dealing with a
6	problem that is not a theoretical or merely an
7	academically interesting problem but one that has
8	real meaning overseas to companies that operate in
9	the global marketplace, which are increasing.
10	CHAIRMAN MAJORAS: A question related to one
11	of Bob's points.
12	The difficulty is we can't know how much the
13	uncertainty contributes to inhibiting procompetitive
14	discounting of price cutting certainly.
15	It is interesting, and Susan will appreciate
16	this. Tim Muris walked into the room and the
17	temperature in my Commission room mysteriously went
18	way down to below levels that I think are
19	appropriate.
20	Susan.
21	MS. CREIGHTON: Representing a lot of
22	high-tech clients --
23	MR. MURIS: I didn't do anything. But I'm
24	warm.
25	MS. CREIGHTON: It is not the thermometer.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

78

1	It is just your presence that causes that.
2	MR. MURIS: I'm not sure what to make of
3	that.
4	MS. CREIGHTON: I can't actually speak to
5	the counseling question you asked, Debbie, because I
6	mostly have high-tech clients, and price bundling
7	isn't a pressing issue so much for them.
8	But I wonder whether some of the problem in
9	bundling isn't so much that this is a huge issue so
10	much as just the LePage's decision was so bad.
11	I would note in the Peace Health case which
12	is one of the ones in the Ninth Circuit, the jury
13	actually found for the defendant in the tying claim,
14	they found no competitive effect.
15	I would throw out the possibility that any
16	reasonable standard amongst whether the AMC or the
17	one that Tim has articulated might go a long way
18	towards addressing the problem.
19	So it is not that you have to get it exactly
20	right than it is the one we have right now is so
21	wrong that it really generates problems that might
22	otherwise be unmanageable.
23	CHAIRMAN MAJORAS: Thank you.
24	Let's move to loyalty discounts and talk
25	about that a little bit. I have a couple of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

79

1	questions that I want to throw out.
2	One is once again looking at what our
3	standard ought to be as we look at this again. This
4	is an area involving price cutting and discounting.
5	So if we are looking at -- when we look at
6	predatory pricing, when we look at bundled
7	discounts, as Bob Pitofsky points out, we have to be
8	careful because discounting is most often
9	pro-consumer.
10	The interesting thing for me when I look at
11	loyalty discounts is to look first at exclusive
12	dealing and the way we look at that. And we find so
13	often that exclusive dealing is not in fact an
14	anticompetitive problem.
15	And loyalty discounts I think, it seems in
16	my mind, then move even closer on the scale toward
17	the area in which we don't have a big problem with
18	it, right, because in many ways, I would think,
19	loyalty discounts are less exclusionary than
20	exclusive dealing, it seems. Yet we do see
21	complaints about loyalty discounts in markets.
22	There is no question about it.
23	First, if you have any views on my general
24	point, and then second, looking at what the test
25	ought to be.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

80

1	I know Professor Hovenkamp has said, as
2	others have, you basically apply a Brooke Group type
3	test to loyalty discounts.
4	What does the group think about what how we
5	ought to be evaluating these situations?
6	Nobody interested in loyalty discounts.
7	MR. MELAMED: No. You were speaking. I was
8	listening.
9	MR. PITOFSKY: I know little about this.
10	Therefore, I will speak on it.
11	I think there is less of a problem with
12	loyalty discounts then with exclusive dealing for
13	two simple reasons. Almost all loyalty discounts I
14	have ever seen are less than 100 percent. They are
15	partial exclusive dealing contracts.
16	Secondly, if halfway through the year you
17	decide it is not worth it, you just opt out of the
18	program. Somebody else comes along and says now for
19	an exclusive dealing contract, I will give you an
20	even better deal, you say, okay, I lose out on my
21	loyalty discount but take your deal.
22	I don't regard it as much of a clog on
23	competition, and it is lowering price in the
24	direction of the consumer.
25	CHAIRMAN MAJORAS: Not a big issue.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

81

1	MR. RULE: Just out of the need to fill some
2	dead air, I again think this is an area where, if
3	you focus on what the exclusionary impact is, it
4	will get rid of a lot of cases.
5	To the extent I have seen loyalty discounts,
6	they tend to have the benefit and they tend to be
7	used with certain distribution channels to incent
8	them to do certain things. It can be a pretty
9	effective tool, at least in theory.
10	The one place where the Department of
11	Justice at least has conducted more than one
12	investigation -- I'm sure they have done it in other
13	places, but the one I'm aware of -- is with respect
14	to travel agent commission overrides in the airline
15	industry.
16	Every time they have looked at them, they
17	have concluded they were not really a problem.
18	One of the reasons they weren't a problem
19	is, first, they were designed to incent travel
20	agents to sell a particular airline's tickets.
21	But, second, by and large, notwithstanding
22	certain articles that have been written by certain
23	people that travel agent commission overrides tended
24	to reinforce hub dominance, the fact is that when
25	you actually looked at the evidence, they weren't

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

82

1	effective in keeping discounters out who came in on
2	a route-by-route basis and basically could get
3	travel agents to sell their tickets on the
4	individual routes as opposed to the network the
5	incumbent carriers had.
6	Generally, I'm not aware of any good case
7	that's ever been pointed to where a loyalty discount
8	has really had an anticompetitive effect.
9	So for that reason, I do think that it is
10	probably not something worth spending a lot of time
11	on. Probably, if you apply a Brooke Group test to
12	it, it will dispose of virtually all of the cases
13	anybody could bring.
14	MS. CREIGHTON: Maybe I could articulate a
15	slightly dissenting view.
16	One of the things that strikes me about
17	loyalty discounts, as compared to exclusive dealing,
18	is they are not found in nature.
19	You find everybody who has exclusive dealing
20	contracts, whether they have 1 percent market share
21	or 50 percent market share. I think we only see
22	loyalty discounts from firms which have substantial
23	positions in the market.
24	I do think it is a question about whether or
25	not in a particular case they can be used to keep

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

83

1	rivals from gaining effective scale.
2	So I think that would be the one context in
3	which I would be interested in knowing more, is
4	whether or not if there are markets in which
5	achieving sufficient scale is critical and the
6	purpose of the loyalty discount really is to
7	foreclose that.
8	MR. MELAMED: I think both of Susan's
9	comments are quite right.
10	But I also think that what Rick said a
11	minute ago is also correct. And that is, if you
12	look at competitive effects, you often can allay the
13	concerns about loyalty discounts because the best
14	theoretical arguments I have heard against loyalty
15	discounts have to do with the steep kind of cliff
16	discount at a particular output, where you are in
17	effect paying a huge discount or sometimes even
18	negative price for the marginal sale.
19	There are many instances in which, if you
20	allocate the discount, as it were, to a handful of
21	sales in order to make the discount look like it is
22	below cost, you will be talking about a volume of
23	sales too small to have an impact on competition.
24	And so, if you marry both Susan's concerns
25	and Rick's focus on competitive effects, I think you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

84

1	still find very few instances in which loyalty
2	discounts are likely to be anticompetitive.
3	CHAIRMAN MAJORAS: Tim?
4	MR. MURIS: The point Susan makes about
5	scale is the modern theory of negative exclusion.
6	But, it has problems.
7	Michael Winston pioneered this theory. In
8	this room on September 11, 2001, unfortunately, we
9	had leading IO economists talking about the issue.
10	Michael said, "it may have helped my reputation, but
11	I don't have a clue if it has any empirical
12	meaning."
13	If what Susan says is correct -- and I don't
14	know that it is or is not -- unlike bundling and
15	exclusive dealing which we find everywhere, loyalty
16	discounts are somehow a practice that we only find
17	with firms with very large market shares, and that
18	would be a very interesting fact. I don't know if
19	somebody has done a survey or has published
20	something. But that would be a fact that would
21	distinguish it from other practices.
22	I still agree with the sentiment that it is
23	hard to think that this kind of pricing practice
24	would be generally anticompetitive. But maybe it is
25	different. I just don't know of that evidence.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

85

1	CHAIRMAN MAJORAS: Okay. Why don't we take
2	a 15-minute break at this point, and we will see you
3	at roughly 11:15.
4	(Recess.)
5	CHAIRMAN MAJORAS: All right. We will get
6	back to it, then.
7	I'm going to start the second half here
8	talking a little bit about tying.
9	We have obviously done some of that
10	naturally in our other discussion, which highlights
11	the fact that it is not very easy to put these in
12	distinct buckets as one might think.
13	Let me just start with a question. There
14	was a lot of discussion on the panels about
15	Jefferson Parish, about the per se rule or maybe you
16	could say the so-called per se rule that the court
17	in Jefferson Parish seems to be laying out there.
18	There was a lot of discussion in our panels
19	about that and I think the belief of a lot of people
20	that in fact they are not even sure that Jefferson
21	Parish really did set out a real per se rule and if
22	it did, that that rule has seen better days and
23	that, in fact, we ought to get on with moving toward
24	admitting that we are moving toward a rule of reason
25	in the tying area.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

86

1	Do folks agree with this? Is this almost
2	without controversy anymore in the United States?
3	Jeff?
4	MR. EISENACH: Yes.
5	CHAIRMAN MAJORAS: That's what I thought.
6	That's why I wanted to get it out of the way.
7	Anybody else?
8	MR. SIDAK: I agree. Uncontroversial.
9	CHAIRMAN MAJORAS: Anybody want to take a
10	dissenting view on that?
11	All right. That's what I thought. We will
12	move on.
13	I want to talk a little bit about something
14	that I find to be more interesting and potentially
15	very important not only in the United States in our
16	dynamic economy today but certainly around the
17	world, and that is tying obviously can be achieved
18	through contract, which is how I think we most often
19	think of it, but it can also be achieved
20	technologically, which we think about more today
21	because the Microsoft case brought it front and
22	center to our attention. But in fact this has been
23	going on forever.
24	Air conditioners, as I understand it --
25	though of course I can't remember this -- used to be

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

87

1	an add-on in your car. I was told this.
2	MR. MURIS: We all know you are a mere
3	child.
4	CHAIRMAN MAJORAS: I wouldn't go that far.
5	But as I was told, air conditioners used to be
6	something you would put in under your dashboard.
7	And eventually the air conditioner became actually
8	part of the car that you buy today.
9	So you could call that, I suppose, a
10	technological tie.
11	Should our standard for legality be
12	different, whether we are talking about contractual
13	tying or technological tying?
14	Greg?
15	MR. SIDAK: I argued since the early '80s
16	that technological tying with respect to product
17	innovations ought to be per se legal, that if you
18	had to choose between per se illegality or per se
19	legality, I think the error costs are such that you
20	are better off not trying to chase this particular
21	business conduct.
22	CHAIRMAN MAJORAS: Susan?
23	MS. CREIGHTON: I'm actually of mixed mind
24	on this.
25	I strongly understand the need to have clear

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

88

1	rules, and I suppose if one -- I can see the strong
2	argument for having a rule of per se legality.
3	I think the only question I have in my mind
4	is if it were shown that the technological tie
5	actually decreased performance of the product, would
6	that cause me to have any different view would be
7	the only reason to tie actually.
8	I don't know. Is this a version of no
9	economic sense? If it actually hampered your
10	ability to sell the product or its performance,
11	would I still be of the same view? And I guess I
12	would throw that out as a question.
13	I'm not sure how I would come out on it.
14	CHAIRMAN MAJORAS: Doug, do you have
15	anything?
16	MR. MELAMED: I understand all the reasons
17	why courts have to tread very carefully in the area
18	of product design innovation. But if a tie or any
19	innovative product design has a tie-out feature,
20	then I don't think we should be talking about per se
21	legality.
22	For example, let's imagine that Microsoft,
23	instead of trying to do in Netscape the way the
24	court found it did, had done it by designing a new
25	operating system that included not only its own

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

89

1	bundler but incompatibility with Netscape,
2	ostensibly because that was the best way to make
3	Explorer work well with the operating system.
4	I don't think that kind of so-called
5	innovation should be beyond the reach of the courts.
6	A test something like Susan articulated would be the
7	right test.
8	MR. SIDAK: Do you think that as a practical
9	matter the outcomes will be much different under the
10	two different rules?
11	MR. MELAMED: The problem is when we talk
12	about practical matter, we are often asking
13	ourselves whether can we think of any cases that
14	would have been decided differently.
15	But if you ask a different question --
16	whether the business community might behave
17	differently -- there is a real risk that a safe
18	harbor for innovation, will induce some firms to
19	manipulate their interfaces and their product
20	designs to exclude nascent rivals.
21	I can't prove that, of course, because we
22	are trying to prove a world which didn't have the
23	deterrent attributes that the law has brought to the
24	world we have experienced. But that would be my
25	conjecture.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

90

1	MS. CREIGHTON: My experience has been
2	counseling on both sides of that question that that
3	kind of arbitrary interface problem actually is
4	rampant in high technology.
5	So I don't think it is actually a
6	hypothetical question. While I'm very sympathetic
7	to the policy concerns about anything less than
8	per se legality, having something less than that
9	could make quite a difference in high technology.
10	MR. RULE: I'm curious, as somebody who
11	occasionally counsels on this issue, how you think
12	that rule would work, Doug.
13	Because it is true that if you have to
14	choose interfaces, sometimes you choose interfaces
15	that, typically you will choose that, allow your
16	products to work better and probably differ from
17	some competitor's product and require the competitor
18	to change its product in order to operate as well.
19	MR. MELAMED: Here's what I would do. I
20	would not do balancing and not do a rule of reason
21	analysis and all that stuff I criticized already
22	this morning.
23	I would say the plaintiff whose product has
24	been excluded by the new design of his dominant
25	rival's product has the burden of proving that the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

91

1	particular aspect or feature or component of the new
2	product that excludes him didn't serve a legitimate
3	purpose.
4	MR. RULE: The problem is that, for example,
5	in choosing interfaces, from what I have seen, to
6	some extent there is an element of arbitrariness or
7	at least subjectivity on the part of the software
8	designer.
9	They have to make choices. And they may
10	make choices that can be viewed objectively by
11	certain engineers -- and, again, the problem with
12	asking an engineer a question is every engineer
13	comes to a problem with his or her own bias. So it
14	is a little hard to ask an engineer.
15	There is that element of arbitrariness and
16	subjectivity. The difficulty is, when you go to a
17	judge, convincing the judge, "well, we had to make a
18	choice at the time, your Honor, this happened to be
19	the sort of technology, the sort of approach that
20	the software designer was used to and preferred, and
21	that's why he or she did it.
22	"But can we say that in some absolute sense
23	it was the absolute best, or that the company spent
24	a lot of time trying to figure out among the
25	different alternatives what was the best or whether

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

92

1	or not instead of coming up with a new version of
2	the interface they ought to just accept either an
3	open standard or some competitor's? No, we didn't
4	do that because that's not the way software is
5	typically designed."
6	MR. MELAMED: In the spirit of the
7	competitor collaboration guidelines, the test is not
8	whether it was the least restrictive alternative.
9	It is sort of ex ante, that, look, it wouldn't be a
10	terrible world, it seems to me, in which dominant
11	firms designing products that exclude rivals have to
12	ask the lawyer can I do this.
13	And the lawyer should say is there a good
14	reason why you are doing it that way, and if there
15	is a good reason, he says it is fine. And if there
16	is not, then maybe you ought to do it a different
17	way.
18	MR. RULE: What if the reason is I have come
19	up with a new innovation that creates value that I
20	would like to capture, and the problem is I want to
21	make sure that I use proprietary interface so I can
22	capture it, so other people can't basically capture
23	it by creating some sort of either peripheral
24	hardware or software that manages to free ride on
25	the efforts that I had? Is there a problem with

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

93

1	that?
2	MR. MELAMED: Certainly appropriating the
3	benefits of innovation, it is a legitimate reason.
4	It depends on the facts.
5	CHAIRMAN MAJORAS: Let's talk about evidence
6	in courts, because we have seen instances in which
7	if jurisdictions show that they are quite open to
8	antitrust claims based on technological issues,
9	based on whether they provide a sufficient interface
10	and so forth, not surprisingly, like bees to honey,
11	the rent-seeking behavior, if you will, the, "well,
12	I want my product to interface on this, this is what
13	my product ought to be able to do with this product"
14	can become quite rampant.
15	Getting down to what are the indicia in any
16	objective sense that the policymakers can look to
17	and ultimately the courts can look to who are not
18	technology experts?
19	What are the factors we would look for if we
20	were going to bring a claim of technological tying?
21	MR. MELAMED: I don't know how to answer
22	that question other than to repeat what I just said.
23	MS. CREIGHTON: I guess I don't see the
24	problem there as being a lot -- certainly from a
25	counseling perspective, it is not a whole lot

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

94

1	trickier in my experience than merger counseling.
2	You say "so why do you guys want to merge?"
3	If they have some plausible story that passes the
4	straight-face test, then you are a lot more
5	comfortable than one who says "the only reason I
6	want to do it is because it excludes my rival."
7	MR. BARNETT: Can I follow up with Susan?
8	What I heard Doug saying, he is not going to
9	balance, that in the spirit of the D.C. Circuit in
10	the Microsoft case, if you have a good reason, it
11	sounded like you were going to call that per se
12	lawful without balancing the potential exclusionary
13	effect of other products.
14	If I have that right, Susan, would you agree
15	with that approach or take a different approach?
16	MS. CREIGHTON: I would agree with that. I
17	actually think the court in Microsoft got it right
18	in the second decision. If you have a plausible
19	efficiency justification, then that would be the end
20	of the inquiry.
21	MR. PITOFSKY: Can I ask a question? I'm
22	with you up until that last point and with Doug,
23	really.
24	Suppose the efficiency is tiny and the
25	anticompetitive effect is substantial. Are you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

95

1	still not going to balance?
2	As long as you can find an efficiency of
3	some magnitude, that's the end of the case?
4	MS. CREIGHTON: This may actually be getting
5	into a can of worms. Certainly in terms of
6	understanding the efficiency justification, unlike
7	Rick, I would want to know whether that is actually
8	why the company did it, as opposed to a post hoc
9	justification.
10	I think if we are talking this little tiny
11	bit and great big anticompetitive effect, I bring a
12	certain skepticism to whether or not the efficiency
13	justification actually is something other than a
14	sort of post hoc rationalization.
15	CHAIRMAN MAJORAS: You really get to part of
16	the point I was hoping we would get to, which is --
17	let me present it as a hypothetical.
18	Suppose we do an investigation and we find
19	all kinds of documents in which a company is saying
20	"I want to do this because I don't want any of these
21	other companies to be able to interface and I want
22	to keep them out."
23	So you get all the sort of bad language, bad
24	intent documents. But then in fact the innovation
25	has proven to be pretty successful for consumers and

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

96

1	consumers like it and it has actually made things a
2	better mousetrap.
3	What do you do with that?
4	You said good reason, bad reason. So they
5	did it for a bad reason, but it turned out to be a
6	pretty good product.
7	MR. MELAMED: I wouldn't focus at least
8	materially what was in their mind, the subjective
9	motive, subjective intent.
10	I think those documents Susan is talking
11	about are very relevant because they can very likely
12	illuminate the underlying economic factors.
13	I would rely on the underlying truth of the
14	matter.
15	Let me add two things. In response to Bob,
16	I actually wouldn't think that just finding
17	something good to be said about the design is
18	enough. In other words, I would ask whether it was
19	really the essential way to design it.
20	Let me tell an anecdote about the Microsoft
21	case. In the Microsoft case, we had on the
22	documents that said Tidalwave and "we have to do
23	something to stop Netscape." And then we had all
24	the conduct.
25	I and others in the Division at the time

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

97

1	said here is all the useful stuff we get from all of
2	our Chicago School defense brief writing over the
3	years.
4	And we served interrogatories on Microsoft
5	and said "why did you do it and where is the
6	compensation that came from that cost?" And they
7	didn't have any answers.
8	Maybe they could have made something up.
9	I'm not sure that the facts play out in quite the
10	stark way that your question suggests.
11	CHAIRMAN MAJORAS: Sure. That's the beauty
12	of hypotheticals.
13	I was about to say I don't even have to turn
14	around and I know who I'm going to next.
15	MR. RULE: Let me tell you the other side of
16	that story, which is actually one of my favorite
17	anecdotes too.
18	I won't necessarily disclose the context in
19	which this came up, and it wasn't Doug asking. By
20	the way, I should just say that I wasn't
21	representing Microsoft at the time those
22	interrogatories were served.
23	But one of the things -- and I think this
24	goes to the question that Debbie posed about what's
25	the evidence.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

98

1	The problem is -- and I don't think
2	Microsoft is that different from what I have seen in
3	other high-tech companies, where you are talking
4	about tens, scores, hundreds, thousands of software
5	engineers developing pretty complex products --
6	It is not really the sort of orderly process
7	that maybe a lot of us lawyers have in mind about
8	how the process works. It tends to be a lot of
9	people working in little collaborative groups over
10	time writing code, then putting it in a tree,
11	compiling it, testing it, going back and writing
12	other things.
13	There is not necessarily a grand scheme
14	every time something is done. So one of the
15	difficulties is that it is very hard to sort of
16	point to a company document that says "here is the
17	strategy, here is why we adopted this, and here is
18	why we didn't adopt that."
19	It is very difficult to think that you are
20	going to find that, at least in a lot of the clients
21	I have seen in the high-tech industry.
22	That brings me to the anecdote. And without
23	disclosing the context, one of the things that
24	somebody who I think is very sensible about
25	antitrust issues, indeed, is generally associated

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

99

1	with the Chicago School, was very troubled by
2	Microsoft's tendency to essentially expend large
3	amounts of money to develop Web-browsing capability
4	within its operating system without having done a
5	cost-benefit analysis before it made those huge
6	investments.
7	This person just could not understand why it
8	was that Microsoft didn't have documents that laid
9	out sort of, "gee, spending $100 million was
10	worthwhile because we could generate this much in
11	return."
12	The fact was -- I don't think Microsoft is
13	that unusual in the real world today when you have a
14	very dynamic economy.
15	What happened was that the company felt --
16	and the Tidalwave document was a good example --
17	that the way computing was moving, it was moving to
18	the Internet, that that was going to be an extremely
19	important function of an operating system, and if
20	you were going to stay current, and if you were
21	going to stay attractive to consumers, you basically
22	had to have that functionality in your operating
23	system.
24	So they didn't take the time to quantify
25	what the costs and benefits were. They basically

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

100

1	said, "we just have to make sure we have that
2	capability in our operating system."
3	I would argue that part of the problem with
4	the like-profit sacrifice test is that the
5	government, and to some extent the courts, took the
6	fact that Microsoft didn't sit down and do a
7	cost-benefit analysis as evidence that, "gee, the
8	only reason they must have done this was basically
9	to put Netscape out of the market."
10	I look at it -- and, again, it is just me --
11	but to me that evidence is equally consistent with
12	the notion that it is a little hard in some economic
13	settings to do a cost-benefit analysis.
14	It made sense to make those investments
15	because the product had to have that functionality
16	if it was going to be acceptable the way they saw
17	the market moving.
18	And they basically said "we don't want to
19	get out of the business, we want to stay in, so we
20	will make the investments that are necessary to do
21	it."
22	To me, that's evidence that that is an
23	efficiency and a justification for the conduct. But
24	the problem with I think some of the tests and the
25	evidentiary rules is the plaintiffs and the court

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

101

1	could look at that same evidence and say, "no, no,
2	that's evidence of profit sacrifice because they
3	were willing to spend anything in order to get that
4	functionality in order to beat Netscape."
5	MR. MURIS: If I could make a historical
6	comment.
7	The context of this discussion about
8	high-tech is so much better than the context 10
9	years ago, which focused on what the evidence showed
10	to be a fallacious view of how network effects
11	made high-tech industries different. Path
12	dependency was said to lead to lock-in and
13	inefficient industries.
14	The claim was based on a couple of examples
15	that turned out to be fallacious, the Qwerty
16	keyboard and on Beta/VHS.
17	The context today here is much more
18	sympathetic to innovation and to high-tech. That is
19	tremendous improvement in a decade.
20	MR. SIDAK: Can I say something about the
21	counterfactual here?
22	We do have some experience with the issue of
23	a large incumbent in a network industry degrading
24	competitor access to the network. It is the
25	telephone industry. It has been subject to heavy

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

102

1	regulation for at least a century.
2	The culture within an incumbent local
3	exchange carrier bears no resemblance to the culture
4	inside of Microsoft.
5	You would not go in to one of the former
6	Bell companies to look for lots of R&D going on.
7	I think the process of subjecting that
8	industry to the degree of regulatory scrutiny over
9	all technical aspects of network interconnection
10	invariably drains it of some of that mojo, if you
11	will, that we hope to see in the computer industry
12	and in other technologically dynamic industries.
13	MS. CREIGHTON: I guess I would have used
14	the telephone industry actually, though, as a
15	counterfactual for why not to have a per se rule.
16	That was, in fact, an industry where there
17	was some technological innovation whose sole purpose
18	was to foreclose competition. So I think --
19	MR. SIDAK: Of what sort? What
20	technological innovation are you thinking of?
21	MS. CREIGHTON: I'm going to get the
22	specific facts wrong. Maybe folks will remember the
23	MCI case better than I do.
24	As I recall, AT&T innovated in a way that
25	required you basically to have these huge boxes that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

103

1	basically would slow down your ability to
2	interconnect with the network. That was an
3	important part of that case, as I recall.
4	MR. SIDAK: I recall the interconnection
5	issues as being a little more pedestrian than
6	inferior access to the network.
7	Why don't we go on.
8	MR. BARNETT: Sure.
9	Given the scarce resource of time, why don't
10	we move on to our next topic, which has to do with
11	refusals to deal with a rival.
12	I guess this has some connection to the
13	telecommunications industry, at least, for those who
14	have viewed it as having such an application.
15	During the various hearings, there have been
16	a range of views presented. But one of the views
17	suggested that a unilateral unconditional refusal to
18	deal with a rival should not be viewed as an
19	exclusionary act, indeed, should be deemed to be
20	per se lawful under the antitrust laws.
21	Would anyone like to agree or disagree with
22	that statement, that proposition?
23	MR. EISENACH: I will start, and I will tie
24	it directly to the conversation we were just having.
25	If Gillette decides it doesn't want its

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

104

1	razor to be compatible with Bic, independent of
2	technological tying, as it were, why can't it just
3	say no, in the same way that Verizon can just say
4	no?
5	I think the issue here goes very quickly to
6	the question of the cost of the alternative, or the
7	"catching the fire engine" problem.
8	Obviously, the European Union is dealing in
9	a much different way with what do you do when you
10	catch the Microsoft fire engine than the United
11	States did. That was always the problem.
12	What do you do when you catch the
13	technological tying fire engine, or what do you do
14	when you catch Verizon?
15	What we have done with the telephone
16	companies in the U.S. is impose a stultifying
17	regulatory regime which very clearly, and I think
18	unambiguously now in the economic literature has
19	been shown to have, resulted in the kind of
20	competition that Scalia talked about in Iowa
21	utilities, which is competition not at the point
22	where innovation occurs and not at the point where
23	costs can be reduced. And at the same time it has
24	dramatically reduced innovation and investment at
25	the core of the network where real competition now

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

105

1	finally is developing in the U.S. about six or seven
2	years after we began removing the worst of the
3	regulatory regime.
4	I think the problem in both cases is that
5	the remedy probably is worse than the disease.
6	If I own the only well, I guess I feel like
7	you have to demonstrate to me that there is no other
8	well possible before I start thinking that the
9	benefits of regulating access to the well exceed the
10	costs.
11	MR. BARNETT: Following up on that, the
12	question is should it be per se lawful without
13	regard to whether or not there is another well.
14	And I guess a related question is are you
15	saying if we may compel some sort of dealing in
16	unique circumstances, should we do it through
17	antitrust laws or separately through regulation?
18	MR. EISENACH: I think the history of
19	innovation has shown there is almost always another
20	way, other than regulation, to skin that economic
21	cat.
22	And the flip side is that when that isn't
23	the case, the cure is often worse that the disease.
24	Again, I think the Europeans' experience with
25	Microsoft is as bad as our experience has been with

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

106

1	trying to regulate telephone companies.
2	The Europeans' experience with Microsoft
3	shows that there is a worse way to do it, and they
4	found it.
5	MR. BARNETT: Bob?
6	MR. PITOFSKY: This is going too genially
7	here. I think I will stir things up.
8	Let me start by saying that mandated dealing
9	by a single firm, even a monopolist, with applicants
10	should be very rare. It just doesn't come up all
11	that often. But I'm not comfortable with never.
12	I think, like the discussion of Section 2, I
13	think a balancing test, of the kind put forward by
14	the Supreme Court in Aspen, is the way to go.
15	There was nothing good about denying the
16	four-mountain ticket in Aspen. And the evidence was
17	that consumers preferred it. So it was a
18	pro-consumer effect that was cut off for no good
19	reason.
20	The problem is -- and I know if I don't say
21	it right now, others will leap in -- what is the
22	remedy? Can you get to a remedy that makes sense
23	and doesn't use the same phrase I used earlier, do
24	more harm than good?
25	And if that's the case, then we have no

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

107

1	right to impose on companies a remedy that we can't
2	describe and we can't enforce and they can't abide
3	by.
4	But I think the difficulties in getting to a
5	remedy have been exaggerated.
6	Take Aspen. They were licensing other
7	mountains in other parts of the west. Then all of a
8	sudden, they go over to Aspen and they cut somebody
9	off abruptly with no reason.
10	I don't think the remedy is very difficult.
11	You take whatever the arrangement was in the other
12	resort areas and apply it to Aspen.
13	There is a question if in the presence of a
14	regulatory agency, is it easier to impose a remedy.
15	And I remember Phil Aveeda making quite a point of
16	the fact that Otter Tail was an extreme case, but
17	the Federal Power Commission was available to handle
18	the details of the remedy.
19	Third, what the Europeans do is send the
20	parties into a room and say "negotiate, come up with
21	something, and if you don't, we will have mandatory
22	arbitration."
23	Imposing a remedy is very difficult. If it
24	is impossible, then the government shouldn't be in
25	it.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

108

1	The point about -- I have a well, and before
2	I think access should be mandated, I want to know
3	that there is no other well there. Absolutely
4	right. That's the point of "essential."
5	If it is not an essential facility, there is
6	no reason for the government to intervene.
7	But if it is, then the question is can you
8	have an essential facility doctrine, as I believe is
9	the case in most countries developing antitrust law
10	in the world, Europe, China and elsewhere, a narrow,
11	narrow, narrow, remedy?
12	Are we disserving antitrust purposes? I
13	don't think so. Certainly I think the lower
14	courts -- I think MCI is the best case for setting
15	up a whole series of conditions before you get
16	access to an essential facility -- sensibly take the
17	remedy question into account.
18	I do not think that unilateral refusal to
19	deal is per se legal. Close to it, but not there.
20	CHAIRMAN MAJORAS: Of course, I would add
21	that sometimes we do the negotiation thing in the
22	U.S. too.
23	Judge Kollar-Kotelly forced Rick Rule and I
24	into the same room for four straight weeks.
25	MR. RULE: It was very pleasurable.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

109

1	CHAIRMAN MAJORAS: Yes, I enjoyed it as
2	well, Rick.
3	We did come out with something. But I would
4	say that as I look at the implementation of that,
5	which we then stuck Tom with, has been difficult,
6	requiring Microsoft to license server protocols that
7	they had never done before. Whereas, in the Aspen
8	case, yes, they had a history.
9	But where it had never been done before
10	proved to be extremely challenging.
11	We haven't had the problems that the
12	Europeans had.
13	MR. BARNETT: Fair enough.
14	Doug?
15	MR. MELAMED: A couple thoughts. Answering
16	the liability question with the remedy question is a
17	mistake.
18	We prohibit murder even though we can't
19	resurrect the corpse. It may be the solution is not
20	to have equitable remedies where we try to regulate
21	the market but, rather, to have a deterrent in the
22	form of exposure to treble damage fines.
23	I think we ought to separate the issues of
24	if there is a disease versus is the cure going to be
25	worse.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

110

1	A couple thoughts on what the rule ought to
2	be. Trinko, by the way, wasn't really about
3	dealing. It was about divesting a very peculiar
4	circumstance there.
5	As a practical matter, we don't need to make
6	it a safe harbor or per se lawful because it will be
7	a very rare case, as experience has shown.
8	He has to have a benchmark. If you don't go
9	in and say you want it for nothing, you have to say
10	he wouldn't sell it to me at price X. The terms are
11	these.
12	It is going to be very hard for a plaintiff
13	to win a case without a contemporary discriminating
14	benchmark.
15	Having said that, we ought not to have a
16	per se lawful rule because when an AT&T refuses to
17	deal with a rival even though it deals with others
18	interconnecting into the market or when an Aspen
19	refuses to accept tickets sold at retail prices to a
20	competitor, there ought to be some room to say now
21	we know he has gone too far.
22	MR. RULE: Let me make two points. It seems
23	to me that one of the reasons -- and I obviously
24	will come to this -- why liability and remedy are, I
25	think, kind of unacceptable is, if you can't think

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

111

1	of an equitable remedy, there may be reasons you
2	don't want to impose it.
3	But if you can't think of an equitable
4	remedy, which is to say a rule, it may suggest that
5	there is some at least fussiness around what you are
6	telling a defendant to do.
7	The problem with your analogy to murder is
8	it is easy to enunciate the rule to society, "don't
9	kill other people," and it may be that you can't
10	resurrect the dead, but you can certainly impose
11	punishments to deter future folks from engaging in
12	that conduct. That is a very clear rule.
13	MR. MELAMED: I have a rule. It is don't
14	refuse to deal when it wouldn't make sense.
15	MR. RULE: If you have a rule that says
16	don't refuse to deal without the when, I could
17	understand.
18	The problem is, it seems to me, once you
19	acknowledge that you have the when, if you have the
20	condition, and then if you add on to that what I
21	think both you and Bob have said is that it is a
22	very rare case that you would ever want to impose
23	some liability for that, it seems to me there is a
24	very strong argument for a rule of per se legality.
25	It is false, it seems to me, to say that,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

112

1	"gee, you can only have a per se rule of legality
2	when you know that in 100 percent of the
3	circumstances the activity is not going to harm
4	competition."
5	That's not the reason that you have a per se
6	rule. Because, you can't even say that in 100
7	percent of the cases of price fixing that there is
8	going to be harm to competition.
9	That's not the reason we have a per se rule.
10	We have it because of error costs.
11	It seems to me that in the area of refusals
12	to deal, particularly if you are talking about
13	unconditional unilateral refusals to deal, the
14	circumstances under which you would ever be
15	concerned about it are so limited and so rare that
16	that's precisely the kind of place you would want to
17	have a rule of per se legality, if for no other
18	reason than saving the courts and the enforcers
19	resources that are otherwise expended investigating
20	and potentially looking for the needle in the
21	worldwide haystack.
22	MR. BARNETT: Tim?
23	MR. MURIS: I like the somewhat Delphic
24	statement in your very good report that came out
25	recently about how it has no meaningful role in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

113

1	antitrust.
2	Let me also say a word about Aspen and ask
3	Doug a question. What the Supreme Court did, given
4	the posture of the case before it, made sense. But,
5	the reality of the case is a business dispute about
6	sharing the profits.
7	MR. MELAMED: It made no sense. I agree.
8	MR. MURIS: Suppose it came to the court
9	that way. Is that a legitimate business reason?
10	Of course, it was a forced bargaining
11	situation, and we know what often happens in forced
12	bargaining situations. You know how they resolve
13	the dispute? They merged.
14	Suppose that had been the context, that
15	Aspen said, "These guys are being unreasonable, and
16	we think we are not getting a big enough share of
17	the profits?"
18	MR. MELAMED: I haven't actually thought
19	through precisely how that would play out. The case
20	was presented in a very odd way.
21	MR. MURIS: And there obviously wasn't a
22	market.
23	MR. MELAMED: Fair enough. So it changes
24	the effects.
25	MR. SIDAK: Could I add a point here about

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

114

1	price?
2	In sector-specific regulation, call it the
3	access pricing problem. There is no problem with
4	granting a competitor access to your facility if you
5	can agree on prices, terms and conditions that are
6	mutually acceptable.
7	The problem is the incumbent will always say
8	you are not compensating for the opportunity cost of
9	the asset.
10	So the access seeker then tries to invoke an
11	antitrust remedy or a regulatory remedy or an
12	arbitration remedy, in the hope of getting a price
13	that's closer to the incremental cost.
14	Is that a problem? Well, it depends on your
15	perspective.
16	If the network only exists because of a very
17	large expenditure of sunk costs, there has to be
18	some contribution to the recovery of those costs
19	beyond the incremental cost of the use of the
20	network.
21	That's what the whole decade of litigation
22	over the Telecom Act in 1996 was all about. They
23	get you into the question of regulating price, which
24	is fundamentally not something that a court can do.
25	It is not even clear that constitutionally

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

115

1	they should be doing because the Supreme Court tells
2	us that price regulation is a legislative function.
3	MR. MELAMED: What he is not entitled to is
4	to refuse a price that is equitable for the purpose
5	of gaining additional market power in some adjacent
6	market.
7	I realize this is very difficult for a
8	factfinder to prove in the absence of
9	contemporaneous discrimination as a benchmark.
10	But what if we could stipulate that the
11	defendant refused to deal on a price equal to his
12	opportunity cost and did so as part of a longterm
13	strategy to preserve or gain market power in an
14	adjacent market?
15	MR. SIDAK: It is plausible. But basically
16	then you are talking about a kind of predation
17	strategy.
18	MR. MELAMED: Yes, one that made no economic
19	sense but for the extra market power.
20	MR. EISENACH: This is one where type 1 and
21	type 2 errors matter tremendously.
22	The reason you have per se rules is not
23	because you are 100 percent sure but because the
24	cost of error is so high.
25	You don't get a second well. That's the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

116

1	cost. The cost of regulating the telephone sector
2	in the U.S. was we didn't get a second network.
3	It is called cable and we now have the most
4	competitive telecom sector in the world as a result
5	of removing excessive regulation. And we are now
6	getting wireless. But that is all because the lack
7	of the regulatory remedy, taking away the regulatory
8	route to a free ride on the incumbent's network.
9	The problem in all this is I don't know how
10	you find the opportunity cost of digging the well.
11	Maybe he kept records of how long he was there with
12	the shovel.
13	But trying to find the opportunity cost of
14	the telephone network is a problem.
15	MR. PITOFSKY: I have been waiting to ask
16	this question for quite some time.
17	What is the empirical evidence, not the
18	theory, empirical evidence, that a mandatory
19	requirement that you deal or you disclose
20	information to rivals is going to lead to a
21	reduction in innovation or a reduction in people
22	coming in and digging a second well?
23	MR. SIDAK: In England, the cable industry
24	vigorously opposed greater unbundling obligations
25	placed on British Telecom, precisely because it

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

117

1	destroyed their business model.
2	MR. PITOFSKY: What did they rely on?
3	MR. SIDAK: Their own wires.
4	MR. PITOFSKY: No. What empirical evidence
5	did they rely on that this remedy would do harm
6	because it would raise barriers to entry to new
7	people who would come into the market?
8	MR. SIDAK: They were in the market at that
9	point, and they were making decisions about
10	investment over time, sequential sunk investment.
11	So it is not really -- in their case, it
12	would not be a question of is there some third party
13	who will enter but, rather, will I currently, a
14	competitor of the incumbent firm, continue to invest
15	in expanding my network or will I simply stop
16	investing.
17	MR. PITOFSKY: I don't want to limit this to
18	telecom. I guess I'm trying to make a very general
19	point.
20	I am upset with the following process of
21	thinking. This is a very, very difficult issue and
22	the remedy is extremely difficult to work out and,
23	therefore, let's call it per se legal. I don't
24	think that's the way antitrust law should proceed.
25	MR. RULE: Bob, you have to add to that the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

118

1	question of its frequency and the question of in the
2	instances where folks have gone after it, whether
3	you think there have been greater incidents of false
4	positives versus false negatives and what the cost
5	is of going after it.
6	I think the frequency is important.
7	Whatever you want to say about the one well, there
8	aren't very many one-well situations in the world.
9	MR. PITOFSKY: I agree with you. I'm with
10	you.
11	I'm sorry. I should have elaborated on this
12	point.
13	I think you have to talk, you have to look
14	at free riders, false positives, false negatives.
15	But I want to do it on the basis of empirical data
16	and not on theoretical assumptions.
17	MS. CREIGHTON: I just wait to ask a
18	question. I don't know this. I thought Bill
19	Kolasky's comments, Doug's partner, were quite
20	interesting at the hearing on refusals to deal.
21	He was articulating how he thought a sort of
22	step-wise application of the Microsoft test would
23	work quite well here.
24	But he observed I think that in the cases
25	where there have been problems, either MCI, AT&T or

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

119

1	Otter Tail, it was part of an overall course of
2	conduct, which I thought was an interesting
3	observation. I would also note in both those cases
4	there wasn't a regulatory overlay.
5	Again, I would just pose the question
6	whether or not that combination of factors calls for
7	sort of a potentially different inquiry, and then if
8	we look overseas, whether they are likely to find
9	that combination of factors more often than you
10	would here in the United States and how the
11	articulation of a rule of per se legality would
12	maybe not be helpful in advancing the analytical
13	debate worldwide about how those issues should be
14	addressed.
15	MR. RULE: Can I make a comment on that?
16	I would take the opposite view. To the
17	extent that the United States equivocates because of
18	penumbras and says we don't think we can have a
19	per se rule of legality, because there may be some
20	incident where there is a problem. And the two that
21	you mentioned and, frankly, the ones that sort of
22	classically I have always thought about, I would
23	argue frankly are as much a function of the
24	regulatory regime that was in place, as opposed to
25	anything that you would have seen in the absence of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

120

1	the regulatory regime.
2	I think that is kind of Jeff's point.
3	I think I will grant you that if you take a
4	position that unilateral unconditional refusals to
5	deal are per se lawful, that will be a somewhat
6	controversial position outside the United States.
7	But on the other hand, I would say that the
8	United States would be in a better position to make
9	certain arguments because I think there is a sound,
10	logical, and I think also empirical basis for taking
11	that position, and taking it and taking a stand on
12	it, and arguing and explaining why that's a
13	reasonable rule.
14	Once you start adding in the equivocation,
15	we may all -- Bill Baxter used to have this saying,
16	that if he got to make all the decisions, he would
17	be fine with basically everything being potentially
18	subject to antitrust regulation.
19	His concern was that he wasn't going to get
20	to make all the decisions.
21	The same thing is true in the United States.
22	In our hands, sort of an equivocal rule may be okay
23	because we are smart enough, sophisticated enough to
24	figure out how to work it.
25	I always worry if you have an equivocal rule

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

121

1	like that in the hands of others whose incentives
2	may not be as pure, whose training and experience
3	may not be as exemplar as our heads of agencies,
4	that in effect they are going to abuse that
5	equivocation in a way that's very harmful to the
6	economy.
7	I think there is at least some argument that
8	they have already done that.
9	MS. CREIGHTON: I guess I was responding to
10	your point in rejoinder to Bob, which was the reason
11	for saying never, not seldom, was because it is
12	rare.
13	I'm just asking if then our articulation of
14	why our answer is never and not seldom doesn't
15	resonate with the experience of folks elsewhere,
16	whether that is maybe not the strongest basis on
17	which to articulate the rule.
18	MR. MELAMED: Let me say relating to that
19	the question, of course, is not is it rare but would
20	it be rare if we had the rule of per se legality?
21	MR. PITOFSKY: Would it be so rare if in
22	fact it became per se legal?
23	MR. RULE: I think you can ask the question
24	a little bit differently. Jeff's question to some
25	extent is the reasonable one.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

122

1	How many single wells are there, how many
2	truly essential assets are there that can't be
3	duplicated that we don't want to be implicated in
4	some way? That's really the issue, I think.
5	I don't think that you can look at the
6	economy and say there are a large number of
7	incidents of those kinds of assets.
8	I can say that there are a much larger
9	number of cases where plaintiffs have argued that
10	there are single wells when there really aren't.
11	That's the danger.
12	MR. PITOFSKY: You can distinguish those
13	cases on the record. You say that only one well can
14	be built here. If it is obvious there can be two,
15	you lose your case.
16	MR. RULE: But it is not costless to do
17	that.
18	MR. PITOFSKY: Of course it is not. We can
19	call everything per se legal and save a lot of
20	costs.
21	MR. RULE: That's not the point. The point
22	is that you could say that there is no such thing as
23	per se illegality because there are times where you
24	could prove that a price-fixing agreement doesn't
25	harm competition.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

123

1	We don't do that because we don't want to
2	expend the resources to try to distinguish those
3	situations.
4	I don't understand logically why the
5	converse doesn't apply as well with respect to
6	conduct that you expect to be so rare and the cost
7	of finding those that are actually problematic are
8	so high that under those circumstances you decide
9	you have a rule of per se legality, recognizing that
10	some harm may go unpunished.
11	MR. PITOFSKY: It won't be so rare when it
12	becomes per se legal.
13	Let me ask you a question. It is exam time.
14	I can't help it.
15	I gather that your approach would overrule
16	Aspen, overrule Otter Tail. My question is would
17	you also overrule Lorain Journal, which was a
18	refusal to deal?
19	MR. RULE: I'll be honest. I'm not a big
20	fan of Lorain Journal. I have said that on a number
21	of occasions.
22	Part of the problem I have with it -- it is
23	a different issue, to some extent. The problem I
24	have always had with Lorain Journal is it doesn't
25	look at the competitive impact that conduct had, in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

124

1	my opinion. It is sort of the precursor of a lot of
2	the unitary tests. I'm not a big fan of it.
3	MR. PITOFSKY: I am a great fan of Lorain
4	Journal. It is the most extreme case I know of
5	where there was no justification and there was a
6	significant anticompetitive effect. This side of
7	the scale had nothing on it.
8	MR. BARNETT: With that, I hope you won't
9	take this as a refusal to deal with the issue
10	further, but I will suggest that we move on to cheap
11	exclusion.
12	CHAIRMAN MAJORAS: I will talk briefly about
13	cheap exclusion. Then we have two more important
14	topics to cover.
15	The Court of Appeals in Microsoft in 2001 in
16	upholding Microsoft's liability did so in part on
17	the basis of an act of deception that it found --
18	that the trial court found Microsoft engaged in.
19	The Commission in its Rambus case used
20	similar conduct in finding Section 2 liability.
21	Is there anyone here who does not agree that
22	misleading or deceptive conduct could be considered
23	to be exclusionary conduct under Section 2?
24	And if it can be, how would others draw the
25	line between situations that justify antitrust

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

125

1	involvement and situations where you might say,
2	well, there is a contractual problem here or perhaps
3	a tort problem, but we don't see an antitrust
4	problem?
5	Doug, do you want to? Moving to another
6	case.
7	MR. MELAMED: I think that conduct that is
8	misleading or deceptive can be anticompetitive
9	conduct.
10	Microsoft Conwood -- and logic make that
11	clear. But it is not anticompetitive conduct
12	because it is susceptible of being labeled
13	misleading or deceptive.
14	Trinko made clear that conduct that is a
15	breach of contract and indeed conduct that violates
16	nonantitrust federal law, is not exclusionary or
17	anticompetitive conduct for antitrust purposes.
18	It seems to me that the Court in Trinko was
19	completely right in that. The issue is does it
20	violate and run afoul of some proper antitrust
21	standard. Yes, causation and all that have to be
22	satisfied.
23	One more brief thing, cheap exclusion.
24	Susan's paper I think on that is a wonderful,
25	insightful contribution to our understanding of the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

126

1	world. It is a very intelligent elaboration, it
2	seems to me, of the Chicago School insight that
3	predatory pricing is an unlikely strategy because it
4	is so costly to the defendant.
5	It points enforcers and plaintiffs in the
6	direction of conduct that is more likely to be
7	mischievous.
8	I don't think it is a concept that helps us
9	answer the question we have been talking about today
10	because as I understand the paper, it identifies a
11	category of conduct that one is cheaper and
12	therefore we should suspect the defendants might
13	want to engage in it. Two, it has no legitimate
14	purpose.
15	I think that's a subset of naked exclusion
16	and with the other elements, market power and all
17	that proven, seems trivial to say that's an
18	antitrust violation.
19	Labeling it deceptive doesn't really advance
20	the question of whether it is anticompetitive. That
21	depends on how it measures up against the
22	preexisting antitrust test.
23	CHAIRMAN MAJORAS: Tim?
24	MR. MURIS: Viewed another way, and this is
25	hardly a declaration against interest I'm making

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

127

1	here -- cheap exclusion is an extraordinarily useful
2	way for the government to think about
3	monopolization. In Susan's phrase it means fishing
4	where the fish are.
5	If you look at the Bush administration's
6	record on Section 2, I think it is spectacular.
7	There are two settlements that are as important and
8	as large as any in history in terms of their
9	monetary relief to consumers, Unocal and BMS, where
10	the FTC worked with the states. By focusing on
11	fishing where the fish are, you are much more likely
12	to produce benefits for consumers and thus have the
13	record of the last several years.
14	So in that sense, which is different than
15	the previous discussion, it is where the government
16	ought to put its effort.
17	It is an extraordinarily important insight
18	because the history of government in private and
19	Section 2 enforcement has not been a happy history
20	at all. It has been a history mostly of mistakes.
21	The many studies that have looked at cases after the
22	fact have shown that the famous cases, ALCOA, United
23	Shoe, and on and on and on, with rare exceptions,
24	were government mistakes.
25	CHAIRMAN MAJORAS: Jim?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

128

1	MR. RILL: I think there is a risk here of
2	taking an interesting and provocative and I think
3	very logical notion of cheap exclusion and expanding
4	that notion across a variety of practices that may
5	or may not be exclusive at all.
6	It is one thing to suggest that
7	hypothetically someone with an essential patent,
8	truly essential patent, knowingly hides it under the
9	table and manipulates the standard process
10	deliberately to include that patent and then shows
11	up once the standard is adopted and says a-ha, guess
12	what I have, and I'm charging royalties of 50
13	percent of the sales price of the implemented
14	article.
15	That doesn't exist. I'm not involved in
16	Rambus. It doesn't exist very often in the real
17	world, particularly when you are talking about
18	innovative evolutions of highly technological
19	products in a moving process.
20	What is the exclusionary act? Does it
21	require that one engage in a continuing patent
22	search to determine whether the standard evolving is
23	something that relies on the patent? Or vice versa?
24	Does it require some kind of -- I think the
25	issue is related to the remedy here.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

129

1	What is the relationship of a remedy if one
2	is forced then to somehow license the patent to
3	those who want to exploit the standard? On what
4	terms?
5	I think it is no answer to say that the
6	notion of fair and reasonable terms suggests that
7	there has to be some solution ahead of time.
8	I think that the danger of adopting an
9	attractive notion such as cheap exclusion and
10	expanding it across a variety of practices tends to
11	produce possibly oversimplistic results that don't
12	fit in the real world and create serious dangers of
13	overenforcement and inefficiencies.
14	MR. SIDAK: I think the controversy looks a
15	lot like the access pricing problem in network
16	industries in the sense that the objective of the
17	party that is seeking access to the patented
18	technology is to try to get as low a royalty price
19	that it has to pay as possible.
20	It is the same generic problem of whether
21	the incumbent, the owner of the essential patent in
22	this case, is going to recoup quasi rents or not or
23	whether the quasi rents will be extracted by the
24	access seeker.
25	I think it is very, very similar to that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

130

1	problem.
2	CHAIRMAN MAJORAS: Even though it is in the
3	standard-setting context and they can choose a
4	different technology?
5	MR. SIDAK: I think it is a less sympathetic
6	set of facts than the typical network
7	interconnection problem.
8	It is, after all, a contractual
9	relationship. These are repeat-play situations.
10	So there is learning by doing, so to speak,
11	in terms of your negotiation with the community of
12	companies that are involved in the innovation giving
13	rise to this set of patents.
14	Also, I think one of the considerations that
15	is not given enough weight here is due diligence on
16	the part of the parties that find themselves later
17	on in the position of wanting access to the patented
18	technology that they think is being priced too high.
19	These are sophisticated companies. If they
20	were to buy or sell a manufacturing facility, they
21	would expect their lawyers to engage in due
22	diligence for the transaction.
23	Why do we think there should be any lesser
24	degree of due diligence on the part of parties
25	participating in standard-setting organizations?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

131

1	I think the whole characterization of these
2	controversies is such that there is too little
3	consideration given to the amount of precaution, the
4	investment and precaution by other members of the
5	standard-setting organization.
6	CHAIRMAN MAJORAS: I think that's a good
7	point.
8	Of course, there are costs to each
9	individual member going out and getting that
10	information, and some of it may not even be
11	available, which I gather is why standard-setting
12	organizations sometimes put in place rules that say
13	everybody tell us.
14	MR. SIDAK: If you are in a high technology
15	industry investing in trying to resolve uncertainty
16	and plumb the unknown, that's part of what you
17	should be doing, just as what Rick was talking about
18	when Microsoft can't put a price tag on what it is
19	worth to try to be sure that they are around when
20	competition shifts to the Internet.
21	MR. BARNETT: What is the cost, the
22	downside, if you will, from a competition
23	perspective of permitting a standard-setting
24	organization to say rather than us being required to
25	go dig out the weeds, we know you have the answer

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

132

1	and you tell us the answer?
2	Is there a downside from the competitive
3	process?
4	MR. SIDAK: Well, the parties certainly can
5	negotiate over what the degree of disclosure has to
6	be.
7	It seems to me that if the burden is always
8	then placed on some party to inform others, there is
9	a kind of moral hazard problem in that the others
10	don't invest enough in creating their own body of
11	information with which to verify the technology or
12	to explore other technologies that wouldn't put them
13	in a bind later on.
14	It seems to me that it sounds good ex ante.
15	But ex post, the problem is that somebody will
16	always come back and say there was more that you
17	could have done or disclosed.
18	It is sort of this problem am I my brother's
19	keeper, how much do I have to tell other companies
20	about what I'm thinking?
21	MR. RULE: I think this goes to the last
22	part of Debbie's initial question, which is I don't
23	know the facts.
24	So it may be that what Rambus did was
25	particularly heinous and completely duplicitous or

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

133

1	not. I don't know. But I don't know that that
2	answers the question as to whether or not it should
3	be an antitrust offense.
4	For example, I could certainly imagine an
5	organization that was trying to come up with a
6	standard having all of its members post a bond or
7	enter into some sort of contract that says that they
8	have to make certain disclosures, and there are
9	certain penalties if they don't.
10	To the extent they violate that contract,
11	then there is a contractual remedy. I can also
12	imagine, with respect to a lot of things that I
13	think of when I hear cheap exclusion, that it is
14	fraud or force.
15	Fraud or force is very bad. Generally it is
16	hard to justify it. But there are also a myriad of
17	statutes, tort law, and other things that address
18	it.
19	It has never been clear to me why antitrust
20	needs to come along and sort of compound that.
21	Maybe those other statutes that directly go
22	to that sort of conduct, frankly, particularly since
23	that sort of conduct is generally going to be bad
24	regardless of the market power or potential market
25	power of the person exercising it, it seems to me

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

134

1	that maybe leaving it to those other statutes is a
2	better way to go than trying to import it into
3	Section 2.
4	MR. EISENACH: Just to frame what you just
5	said, it is the equivalent of burglary with and
6	burglary without a gun or armed versus unarmed
7	robbery.
8	What we are saying is, the act performed
9	outside the context of an anticompetitive scheme
10	gets a penalty. The act performed in the context of
11	an anticompetitive scheme gets a triple penalty.
12	MS. CREIGHTON: I guess I would turn that
13	around and say in criminal antitrust, I don't think
14	we would say we will only apply the criminal
15	antitrust statutes unless we first find that the
16	conduct isn't reachable by mail and wire fraud.
17	I think it is a separate and independent
18	question. I think whether it is a tort, not all
19	torts are antitrust violations, and obviously most
20	antitrust violations aren't torts.
21	But I don't think we would want to say
22	because it is a tort that therefore something that
23	otherwise would be an antitrust violation therefore
24	on that ground alone should be immune.
25	MR. RULE: I think -- not disagreeing

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

135

1	necessarily with what you are saying, but I will say
2	that I think the analogy to criminal law is wrong.
3	Because the only reason -- I think there is a good
4	basis for saying this --
5	The only reason that the conduct that is
6	also challenged as wire fraud or mail fraud is
7	challengeable is generally because the underlying
8	conduct violates the antitrust laws for various
9	reasons.
10	There are certain exceptions and certain
11	times that you can challenge it as an attempted wire
12	fraud, whereas, you couldn't challenge it under the
13	antitrust laws.
14	It is because the underlying act itself
15	would violate the antitrust law.
16	My only point is there are certain downsides
17	to Section 2 enforcement, including whether the
18	penalty -- I guess you could say that for a lot of
19	cheap exclusion, because it has no socially
20	redeeming value and we can always identify it
21	perfectly, who cares what the penalties are.
22	But to the extent that's not the case, and
23	to the extent there are other regimes that are
24	intended to impose punishments and they are optimal,
25	then adding antitrust on top of it, to me at least,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

136

1	arguably creates suboptimal enforcement because you
2	have too much enforcement.
3	MR. BARNETT: Okay. I want to move quickly
4	to the international setting and make sure that we
5	leave time for remedies as well, which I think a
6	number of folks think is a very important topic.
7	On the international fronts, let me ask
8	Bob -- I will start off with you, if that's okay --
9	whether there are particular areas that you are
10	aware of where there is not currently convergence
11	between the United States and other jurisdictions
12	around the world in terms of unilateral conduct
13	enforcement.
14	And a related question with respect to
15	those, presumably we should be trying to move
16	towards some convergence, would you rather see
17	convergence for its sake or only if it goes in one
18	direction, the right direction, if you follow?
19	MR. PITOFSKY: I can go on for a long time
20	about where divergence is occurring.
21	I just finished teaching a seminar on
22	comparative antitrust. I will just pick two.
23	Others will probably want to add different
24	examples. Dominant firm behavior is diverging, not
25	just between the United States and Europe but

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

137

1	between the United States and many other countries.
2	Second, I'm not sure there ever was
3	convergence, but the United States' position that
4	only economics matters and no other factors should
5	be taken into account is practically unique in the
6	world.
7	I'm not sure we are wrong about that. But I
8	would simply point out that there are 104 countries,
9	and 103 of them don't seem to be going along with
10	that kind of approach.
11	Is convergence a good idea? Yes, I think it
12	is. I think we are going to get more. We have had
13	quite a bit already.
14	Just take EU and U.S. definition of relevant
15	market, attention to distribution arrangements, oh,
16	and worldwide, worldwide agreement that cartels do
17	no good, and they ought to be challenged in the most
18	vigorous, serious way.
19	So you do have convergence. I think
20	convergence is a good idea. We ought to achieve
21	more of it.
22	I hope I will have a chance to talk about
23	comity later on.
24	CHAIRMAN MAJORAS: Jim?
25	MR. RILL: I think we should not be too

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

138

1	pessimistic and certainly not too humble about the
2	opportunities for convergence and the role the U.S.
3	should play.
4	There is an enormous track record with
5	respect to merger enforcement and cartel
6	enforcement.
7	And I think possibly we can see at least
8	through the discussion draft some move on the part
9	of the European Union, coming more under the
10	discussion draft, towards looking at an
11	effects-based analysis under Article 82.
12	I think the role of the United States is
13	critically important in its maturity and development
14	that it has contributed to antitrust.
15	I think sometimes we are criticized and more
16	often we criticize ourselves for saying convergence
17	means do it our way. That's not the case.
18	I think we do somehow, I think, get an
19	attack made on our credibility by those who say you
20	don't bring these kinds of cases, why should you
21	tell us not to bring these kinds of cases.
22	I don't think we tell the story that I think
23	Tim was talking about and Justice could say as well
24	that we have brought the right kind of cases. Some
25	might argue whether they are.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

139

1	We have a story to tell here. There is an
2	economic basis that needs to be explained and I
3	think there is an opportunity for progress there.
4	Without prolonging it, there are
5	organizations and institutions for that progress to
6	be made through the ICN and OECD through the
7	cooperation that has been developed. I think there
8	is much to be done in the area, and it shouldn't be
9	abandoned with respect to Section 2, Article 82 and
10	whatever is going on in the Far East.
11	MR. BARNETT: Rick, anything?
12	MR. RULE: The only thing I would say is if
13	given the choice between convergence and advocating
14	what you believe is the right principle, I would
15	frankly urge you always to adopt the second.
16	I think that ultimately convergence is
17	important, and the fact that there is divergence in
18	certain areas can be very costly and painful to some
19	companies. And I think that in terms of cost,
20	obviously convergence is a good thing.
21	The problem is if you compromise in terms of
22	your position, and I think that even though
23	obviously I have some disagreements with where U.S.
24	positions have evolved, the fact is they are backed
25	up by a lot of experience, and I think they are

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

140

1	pretty sophisticated.
2	By and large, they are the most defensible
3	positions. I think it is a mistake if in the name
4	of convergence you move away from the right
5	principles.
6	If you advocate the principled position and
7	explain why it is the principled position, even if
8	people won't accept it today, they will accept it
9	later. The one example I will give is cartel
10	enforcement.
11	When I was in the Department of Justice, we
12	were ordered by the President to shut down an
13	investigation of airline price fixing over the
14	Atlantic by the British government, which called us
15	a banana republic for criminally enforcing antitrust
16	laws.
17	Well, guess what --
18	MR. MURIS: Actually, I was there. He said
19	they were acting like a banana republic.
20	MR. RULE: Yes. But they were calling us
21	other names.
22	But what the United States did was stick to
23	its guns, that cartel behavior is bad, severe
24	penalties are appropriate to deter the conduct, and
25	over the course of 25 years, it has actually brought

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

141

1	the rest of the world around.
2	It is important to maintain the principled
3	position, and ultimately people will follow you.
4	MS. CREIGHTON: In that regard, I could be a
5	one-trick pony here. We don't have to call it cheap
6	exclusion. We could just call it naked exclusion.
7	If you think internationally there is some
8	benefit to culling out, that the agency should focus
9	on instances where they know there is competitive
10	effects and there is no cognizable efficiency
11	justification, that if they are going to have civil
12	nonmerger investigations, that's where they should
13	focus, just like we have told them in mergers, it's
14	good to focus on horizontal mergers, not vertical
15	mergers, it is good to focus on cartel behavior,
16	because it has a much less kind of chilling effect.
17	And I think I probably disagree with Jim a
18	little bit in that what Tim was saying was that if
19	you view naked exclusion as Doug had defined it, as
20	reducing the output of your rivals so that it
21	crosses both Sections 1 and 2, virtually all of the
22	FTC's real estate cases sort of going all the way
23	back for the last six or seven years, except for
24	Three Tenors, all of them have been instances where
25	it was naked exclusion.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

142

1	One can actually point to that record and
2	say it is actually quite common, it typically
3	involves manipulation of government or
4	quasi-government services, Orange Book cases, for
5	example.
6	That is really where they should be putting
7	their resource dollars, as opposed to focusing on
8	price bundling or refusals to deal with rivals.
9	MR. BARNETT: Doug?
10	MR. MELAMED: I agree with almost everything
11	that Susan said. But I don't agree with the
12	implicit characterization of Rambus as a case of
13	naked exclusion. I guess that is for the courts to
14	decide.
15	Maybe I'm transitioning to the next topic.
16	I want to say the following. I think with time
17	there will be some convergence. Europe doesn't have
18	the treble damages exposure that affects the
19	analysis of false positives and false negatives. I
20	think there will be increasing convergence.
21	In a way more serious than the problem of
22	different substantive rules in different
23	jurisdictions is the problem of overlapping
24	investigation of the same transaction by multiple
25	jurisdictions.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

143

1	It is a problem in the U.S. when the FCC,
2	the states, and the federal antitrust agencies have
3	investigated the same transaction. It is especially
4	a problem internationally with multiple
5	jurisdictions.
6	The problem is not just sort of that there
7	is a search by the complainant for the lowest
8	standard. That is true. It is also that there is a
9	search by the complainant for multiple reviews.
10	Multiple reviews ensure that we are going to
11	have a bias in the system in favor of false
12	positives because the second review can cure a false
13	negative but there is nothing that can cure a false
14	positive.
15	So I think one thing the United States ought
16	to do is to stand firm for the principle that
17	multiple agencies should not be looking at the same
18	transaction.
19	MR. MURIS: Let me make three points.
20	First, I agree with Susan about the
21	empirical significance of cheap exclusion. There
22	has been significant work regarding horizontal
23	activity in this administration. For example, at
24	the Justice Department, grand juries had fallen to a
25	very low level by the end of the last decade. The

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

144

1	Justice Department has built that back up and done a
2	very good job. The FTC obviously did a lot with
3	regard to price fixing and other cases.
4	Second, in response to Bob's point about the
5	United States being unique or close to unique among
6	the developed economies, we are the only one with
7	flexible labor and credit markets, and that is to
8	our enormous benefit.
9	Finally, in response to Doug's point, with
10	which I agree, there is a difference between mergers
11	and dominance. Mergers are divisible in the sense
12	that you can have multiple reviews and it is
13	basically okay because you can sell off parts. But
14	in the dominance area, the most aggressive remedy
15	tends to dominate.
16	MR. MELAMED: I meant to say for the global
17	market situation I agree.
18	CHAIRMAN MAJORAS: Maybe it is a good time
19	to jump in and finish up with remedies.
20	MR. PITOFSKY: Can I say a word about that?
21	CHAIRMAN MAJORAS: Yes.
22	MR. PITOFSKY: I let it go because I wanted
23	to hear what everyone had to say.
24	My view -- I hope it is not too
25	pessimistic -- is that convergence is a long way

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

145

1	off. It seems to be going a little bit the wrong
2	way right now. For example, the WTO has given up on
3	its working group seeking a way to achieve
4	convergence.
5	I commend the other groups that keep at it.
6	And I would keep at it even despite my view that it
7	is not in the cards for the foreseeable future.
8	But I think there is something that is in
9	the cards, and that is comity.
10	And please don't take what I say as
11	deference. I am utterly practical about this. We
12	will never get deference; one country says to the
13	other country "you do it and I will go along with
14	everything you say."
15	What you can have is enhanced comity. This
16	comes back to three or four countries examining the
17	same behavior, and for the second, third and fourth
18	country to say "look, we are going to wait and see,
19	we respect the way you do things, and we are going
20	to wait and see what you do, and if you do the right
21	thing, we will just accept your remedy and we will
22	go away."
23	Canada does it on a regular basis
24	constantly.
25	And I think there are a lot of people around

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

146

1	who believe that it is a step in the right
2	direction.
3	Also, enhanced comity would require that you
4	do everything possible not to have inconsistent
5	judgments, not to say to a company do A and then
6	somebody else says not do A. There are a number of
7	things that can be done here.
8	I regard traditional "comity" up until now
9	as being not frivolous but a trivial matter. It
10	could be changed. It could be changed by treaty.
11	The United States and Europe would get together and
12	offer a program of enhanced comity.
13	I think it would migrate elsewhere. While
14	we are standing around waiting for convergence, I
15	see that as something useful to do.
16	MR. BARNETT: A quick follow-up.
17	If you are going to have this respect, if
18	you will, do you decide who goes first and who sits
19	back and watches?
20	MR. PITOFSKY: Tough one.
21	In the international bankruptcy field, they
22	have that problem. And the answer is that the
23	country that has the most connections with the
24	debtor institution takes the lead. The United
25	States and other countries have committed to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

147

1	deference; anything they say is okay with us.
2	Antitrust comity will be more complicated.
3	But we can find a way to decide which country has
4	the most connections with the transaction.
5	MR. RILL: The principles of traditional
6	comity are spelled out in the US-EU cooperation
7	agreement. Those are taken from a long history of
8	the development of traditional comity.
9	Those elements can lead those who wish to
10	adopt, if you will, a soft deference policy towards
11	a solution as to which country might go first.
12	Whether they can be applied with any degree
13	of comity remains to be seen. The principles are
14	there.
15	CHAIRMAN MAJORAS: Okay. We should move on
16	to remedy, which is an area that is extremely
17	important.
18	I think Doug's point, the D.C. Circuit in
19	Microsoft said yes, it may be hard six years later
20	after the conduct to find the right remedy but you
21	still need to bring the cases because that gives
22	instruction for companies in the future and may have
23	some deterrent value.
24	I agree with all that. But nonetheless, I
25	think the remedy issue is one that has been at least

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

148

1	as vexing as even the liability issue in a number of
2	cases for us.
3	A couple questions I will throw out and
4	folks can jump in.
5	One thing that we looked at was that when we
6	had our panels on remedy, they generally agreed one
7	of the goals should be to restore competition in the
8	market.
9	Is that realistic, particularly in so many
10	markets we deal with today? They are hardly static.
11	You can't pin them down in time.
12	Is that a realistic goal? If so, how should
13	we look at doing it?
14	The other question I would throw out in the
15	interest of time now is if in fact we find that it
16	is very difficult to impose a remedy and, worse yet,
17	imposing a remedy may do more harm than good to the
18	market, then are we better off with doing nothing
19	or, for example, what Doug suggested, maybe having
20	civil penalties, maybe leaving it to treble damages,
21	as opposed to intervening with some sort of conduct
22	or structural remedy?
23	MR. SIDAK: I think that the damages
24	approach has a lot to commend it.
25	If you think about a big case like the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

149

1	Microsoft case, I think it is unfortunate at the
2	very beginning of the case there wasn't a clear
3	statement as to what the desired remedies were on
4	the part of the federal government.
5	The divestiture remedy was something that
6	was introduced publicly at least much later on.
7	The critical issue over which Microsoft and
8	the government disagreed the most was what is the
9	measure of harm to consumers, what is the consumer
10	welfare loss from this.
11	Economists would try to answer that question
12	by measuring damages. It seems to me answering the
13	liability question and getting to an alternative
14	kind of remedy collapses into a single exercise.
15	CHAIRMAN MAJORAS: You should know in the
16	30,000 or something comments we received on the
17	settlement, there were a pretty good percentage of
18	people saying you didn't do your job because you
19	didn't get any civil penalties or damages against
20	Microsoft.
21	Of course, we had no authority to do it.
22	But the general public, when they were doing it
23	without profanity, agreed with you.
24	MR. SIDAK: But there was a prayer for other
25	injunctive relief as the court might grant.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

150

1	I don't know if the court would take that
2	and run with it and cook up some kind of
3	disgorgement remedy perhaps. But I don't believe it
4	was pursued by the Justice Department.
5	CHAIRMAN MAJORAS: Doug?
6	MR. MELAMED: I agree with your bottom line,
7	Greg. But Microsoft illustrates the limits, not the
8	case for money remedies.
9	The Microsoft case at its core was a case
10	about an investment by Microsoft in raising entry
11	barriers.
12	I don't know how you would prove who was
13	damaged by it. There was a prediction that the
14	market would behave less well in the future
15	sufficient to justify the liability determination.
16	If you really thought that the penalty for
17	that should be equal to the damages incurred by some
18	definable body, I'm not sure there would be much of
19	a penalty.
20	MR. SIDAK: That was the problem with the
21	back-of-the-envelope calculations about what would
22	be the profit-maximizing monopoly price for Windows
23	and how does it compare to the price that was
24	actually being charged.
25	To me, that's what was the stark question.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

151

1	Well, if the price that's being charged is so much
2	less than what the profit-maximizing price would be,
3	what is the big problem here, why the big case?
4	MR. EISENACH: Rick may know the number. I
5	don't off the top of my head.
6	I know Sun's damages were I think settled at
7	$4 billion. I don't know what all the totals were.
8	The conduct coming out of that case did
9	translate directly into civil damages on Microsoft.
10	They paid billions of dollars.
11	MR. RULE: They settled with quite a few
12	people.
13	MR. EISENACH: Netscape.
14	MR. RULE: Yes. They did pay. I'm not sure
15	what the number is now, but it may be exceeding
16	$10 billion. I don't know when you add it all up.
17	I guess the thing I would say is it is
18	important for the government to think about the
19	remedy before it brings the case. A lot of people
20	said that.
21	It may not be fair to the government to say
22	they hadn't given that a lot of thought. I don't
23	know. I defer to Doug as to whether or not they had
24	given it a lot of thought.
25	But it does seem to me that the government

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

152

1	ought to have in clear mind what the remedy would
2	be, and if it can't visualize and articulate what
3	that remedy would be, maybe it shouldn't bring the
4	case.
5	Second point --
6	MR. SIDAK: Could you push that a little
7	farther? What should a court do if the government
8	doesn't articulate the remedy in its complaint?
9	MR. RULE: I think it is just the way the
10	rules work. A court is not going to throw you out
11	on that basis.
12	You could establish a rule that says that
13	the remedy ought to be articulated. What a
14	plaintiff and a court may say is we can't fully
15	understand what the appropriate remedy will be until
16	the factual record is built. So you would have some
17	issues there.
18	I think the government doesn't have that
19	problem, and they ought to think about their theory
20	in that way.
21	The second point I would make is, in a way,
22	Microsoft arguably was an easier case. It was an
23	easier case on remedy. I think it ultimately worked
24	out as a remedy because it was a maintenance case as
25	opposed to a monopoly acquisition case.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

153

1	And the core of that remedy was a set of
2	rules that really can be viewed largely as
3	prohibitions. And, frankly, as far as that part of
4	the decree has gone, it has worked pretty darn well.
5	The issues have come up in the Microsoft
6	decree really in two different parts, one that sort
7	of gets into the way the product is designed. But
8	that has generally been fairly manageable because of
9	the way the government ultimately focused in on what
10	the court found to be the problem.
11	The place where there was really the problem
12	was the protocol licensing provisions.
13	I would argue the reason that that was a
14	problem, without going into how it got there,
15	because it really wasn't part of the government's
16	case, and it sort of came in in the course of a
17	negotiation, and it was probably a part of the
18	remedy that was not very well thought out. It
19	didn't really have a basis in the factual record.
20	It has proven as a result to be kind of a
21	difficult one to implement. I think there is some
22	question about how efficacious it was.
23	But I think the strength of that decree is
24	that it enumerated certain practices that had to be
25	proscribed, and it has done a very good job of that.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

154

1	If you can have a decree like that in a
2	Section 2 case, it works.
3	When you start getting into structural
4	remedies, I can see one in a case where a monopoly
5	is created through acquisition so that you have at
6	least arguable natural demarcations where you could
7	divide a company.
8	But otherwise, it is very hard to have a
9	structural remedy. And in my view, that sort of
10	goes back to what Greg is saying. In most Section 2
11	cases, I think the government ought to think about
12	whether it can bring about an appropriate remedy.
13	And where it can't, it ought to recognize there is a
14	treble damage remedy available to the plaintiffs.
15	Plaintiffs in monopoly cases are not somehow
16	fooled. They know where there is a violation. They
17	can go into court. And there is an argument that
18	that is an overdeterrent. But they are entitled to
19	that.
20	The argument would be that it is a rare
21	Section 2 case that the federal government ought to
22	go after and by and large it ought to leave those
23	cases to private plaintiffs.
24	MR. MURIS: The remedy issue I think raises
25	the second major benefit of the cheap exclusion

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

155

1	approach.
2	The first is by fishing where the fish are,
3	you are more likely to find problems. The second is
4	the remedies are generally easy.
5	The remedy that Chairman Majoras proposed in
6	the Unocal case wasn't hard at all. It saved an
7	enormous amount of money for consumers and it
8	involved gasoline, for which the Commission should
9	get a lot more credit than it gets.
10	CHAIRMAN MAJORAS: We get none.
11	MR. MELAMED: The questions that we were
12	given beforehand, the first question on remedies,
13	asked what are the appropriate goals for a Section 2
14	remedy.
15	I think there are six of them. One is
16	general deterrence. Two and three, compensate
17	victims and disgorge profits from the wrongdoer.
18	Four and five, end the wrongful conduct and prevent
19	its recurrence. And the latter includes sometimes
20	fencing in, going beyond the literal conduct. And
21	six is restore competition in the injured market.
22	The first 4-1/2 of these, general
23	deterrence, compensate victims, disgorge profits,
24	end the wrongful conduct, prevent its recurrence,
25	don't raise difficult remedy questions.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

156

1	So what we are talking about here is the
2	scope of fencing-in remedies, which can be
3	problematic and restoring competition in the injured
4	market.
5	It seems to me nutty to say in effect that
6	the complaint should be dismissed if the plaintiff
7	cannot at the outset of the case articulate a
8	coherent remedy that falls into categories 5B and 6.
9	There are too many other reasons to bring a case to
10	be held up on account of an inability to satisfy
11	sensible injunctive remedy.
12	Maybe the answer is many don't have remedies
13	of those types.
14	Two more brief comments. On Microsoft,
15	there were billions of dollars paid largely to
16	existing rivals who claim to have been excluded
17	historically by the antibarrier conduct.
18	That didn't capture the theory of the case,
19	which was consumers and rivals that hadn't arrived
20	on the scene and consumers that would be injured in
21	the future. It is not a bad start.
22	On structural relief, the structural remedy
23	in the Microsoft case, entering it without a hearing
24	was one of the most breathtaking remedies, but
25	conceptually it wasn't a bad idea.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

157

1	If you assume, probably counterfactually,
2	that the ops company and the aps company could have
3	been divided like Siamese conjoined twins without
4	killing one of them without a lot of cost, then this
5	is a remedy perfectly consistent with the theory.
6	It preserves the market power of both
7	companies and changes -- it is a vertical
8	divestiture and changes the anticompetitive
9	behavior.
10	Suppose Office and Windows had been two
11	separate companies? How would we feel about a
12	merger? Might we be very concerned about that?
13	I'm not saying it made sense in fact. I
14	don't think you can say mechanically or
15	formulaically structural remedies are appropriate
16	when you have an illegal horizontal aggregation
17	because they might make the most sense.
18	MR. BARNETT: You sort of blew by your first
19	4-1/2 goals. Can I probe a little bit on that?
20	Take your bundled discount, some of the
21	pricing conduct we talked about before. That puts
22	the court in a position of prohibiting the conduct
23	which I thought was within your first 4-1/2,
24	prohibiting conduct relating to pricing.
25	You view that as a simple, straightforward

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

158

1	proposition?
2	MR. MELAMED: I agree. There may be
3	instances where it gets difficult. Predatory
4	pricing --
5	MR. BARNETT: I didn't give you that
6	example.
7	MR. MELAMED: I understand. What you are
8	doing it seems to me is sneaking in the remedy
9	question an uncertainty about the liability test.
10	Let's take predatory pricing. I don't think
11	we would want to have a remedy that said, defendant,
12	don't sell your widgets for less than $4. But we
13	might say don't sell it for less than whatever we
14	think the appropriate cost measure is and in effect
15	incorporate into an injunction the substantive
16	standard.
17	I think when it comes to a simple sin no
18	more remedy, the difficulties in most cases are
19	going to mirror difficulties in articulating the
20	liability rule. They are not difficulties of
21	remedy. They are not inherent in a remedial scheme.
22	CHAIRMAN MAJORAS: I was surprised to hear
23	you say government should primarily stay out of
24	Section 2 and leave it to the private lawyers, and
25	maybe your view is that is typically business to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

159

1	business cases.
2	I understand. But several of the panelists
3	noted that it was the private treble damage actions
4	that really had the impact in terms of chilling
5	certain types of aggressive behavior and really not
6	the government actions.
7	So someone suggested because it is tough to
8	identify actionable Section 2 behavior, we should in
9	fact take away the treble damage aspect. I know if
10	you did that, I would also have a question, should
11	you give the government civil penalty authority as
12	opposed to a disgorgement situation in the equitable
13	realm?
14	MR. RULE: I will say that my comments
15	before were premised on the assumption that you guys
16	can't do something about the treble damage remedy.
17	I think it would exist. The other
18	qualification which should be implicit in what I
19	said before but I want to make it clear, when I say
20	it should be left to private suits, I mean for
21	damages, not injunctive relief.
22	Because as you know, I think, in my view, to
23	the extent that there are injunctive remedies
24	available to a plaintiff, it probably should be
25	limited to the federal government because there are

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

160

1	too many problems when you expand that.
2	But having said that, look, I think there
3	are definite problems with the treble damage remedy.
4	Twenty years ago, what the Department of
5	Justice suggested was limiting treble damages to
6	suits by suppliers or purchasers, and suits by
7	others who claimed to have been harmed because of
8	lost profits or exclusion from the marketplace would
9	be limited to their actual damages. The argument
10	being that it is not like there is some question of
11	detecting the illegal behavior because you know you
12	are subject to it, so there is no particular reason
13	to give anybody more than compensation for their
14	injury.
15	For that reason, I still believe that
16	probably single damages for competitors who are
17	harmed by that conduct is probably sufficient.
18	But do I think that you guys can bring that
19	about? Do I think Congress is prepared to bring
20	that about? No.
21	I think in light of that, that's why I say
22	since that is going to exist anyway, you may as well
23	leave most of these cases to the private sector.
24	CHAIRMAN MAJORAS: Any other final comments?
25	Any other final comments on anything?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

161

1	We can do like the McLaughlin Group and go
2	down the line.
3	MR. RULE: You have to give us a one to ten
4	question.
5	MR. BARNETT: The benefits of the hungry
6	stomach.
7	MR. EISENACH: I can't resist saying one
8	thing about essential facilities and remedies.
9	The guy who dug the well -- or the guy in
10	Steve Jobs case who created the iPod -- may be the
11	ninth or 10th guy who tried to dig that well. The
12	first nine didn't make it.
13	The probability of a regulatory agency
14	appropriately compensating the 10th guy who finally
15	made it to the bottom of the well and got water for
16	the risk he took is so close to zero to me it just
17	trumps the case.
18	MR. SIDAK: A free option problem.
19	CHAIRMAN MAJORAS: Any last words?
20	Thank you so much, panelists. This has
21	really been tremendous.
22	We thank you for your participation and for
23	taking four hours out of what I know are your busy
24	schedules.
25	MR. BARNETT: I agree. I actually, given

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

162

1	the reputation of the members of this panel, had
2	very high expectation. I'm gratified to say they
3	were exceeded.
4	Thank you.
5	(Applause.)
6	CHAIRMAN MAJORAS: I should probably say
7	this for Pat and Gail. This concludes our Section 2
8	hearings.
9	(Whereupon, the hearing was concluded.)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

163

1	CERTIFICATION OF REPORTER
2	DOCKET/FILE NUMBER: P062106
3	CASE TITLE: SECTION 2 HEARING
4	DATE: MAY 8, 2007
5
6	I HEREBY CERTIFY that the transcript
7	contained herein is a full and accurate transcript
8	of the notes taken by me at the hearing on the above
9	cause before the FEDERAL TRADE COMMISSION to the
10	best of my knowledge and belief.
11
12	DATED: 5/15/2007
13
14
15
16	BRENDA SMONSKEY
17
18	CERTIFICATION OF PROOFREADER
19
20	I HEREBY CERTIFY that I proofread the
21	transcript for accuracy in spelling, hyphenation,
22	punctuation and format.
23
24
25	DIANE QUADE

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

Updated January 2, 2024